Football offers a surprisingly rich archive of strategic thinking. There is a peculiar irony in the story of Osvaldo Zubeldía, the Argentine football tactician who, in the late 1960s, systematically dismantled the romantic notion that football was an art form. His Estudiantes de La Plata side were drilled relentlessly in their training base at City Bell, subjected to 120 different tactical repetitions designed to “exclude coincidence”. They won everything there was to win [1]. Three consecutive Copa Libertadores titles. The Intercontinental Cup. The first professional championship for a “small” club in Argentine history. By any measurable standard, it was an astonishing success. And yet, as Jonathan Wilson notes, the movement Zubeldía created “provoked a profound national neurosis” about football’s very purpose [1].

The parallels to contemporary design culture are not merely illustrative; they are structural. What Zubeldía pioneered in football was a totalizing systems approach that prioritized measurable outcomes over aesthetic expression. The same approach has become the dominant ideology of modern product development, marketing strategy, and organizational design. We have built our own City Bell training facilities, populated them with tactical analysts, and subjected every creative decision to the laboratory method. Design thinking by directive. And like Argentina after 1969, we are discovering that technical virtuosity and joylessness are not mutually exclusive.

Thanks for reading The Spotted Ape! Subscribe for free to receive new posts and support my work.

The Seduction of the System

To understand Zubeldía’s achievement, one must appreciate the context from which it emerged. Argentine football in the 1950s was defined by la nuestra—“our way”—a style predicated on individual brilliance, the gambeta (dribble), and the romantic ideal of the pibe, the street urchin whose natural genius needed no formal instruction [1]. It was beautiful. It was celebrated. And in Helsingborg in 1958, when Czechoslovakia demolished Argentina 6–1 at the World Cup, it was revealed as catastrophically insufficient.

Zubeldía’s response was radical in its clarity: if the bohemian approach had failed, then bohemianism itself must be eliminated. What followed was the first truly modern technocratic intervention in South American football. Players were sequestered at the concentración, isolated from the distractions of family and nightlife. Training became laboratory work: systematic drilling of offside traps, pressing patterns, and set-piece routines until every scenario had been anticipated and optimized [1], [2]. The individual was reconceived not as an artist but as a “functional unit” within a larger system.

It worked brilliantly. Between 1967 and 1970, Estudiantes broke the monopoly of Argentina’s “Big Five” clubs, proving that disciplined systems thinking could overcome superior individual talent [1]. Zubeldía had demonstrated something profound. He showed that football, or indeed any complex endeavor, could be approached as an engineering problem, decomposed into its constituent elements, optimized through repetition, and executed with mechanical precision.

The design world has spent the last two decades learning precisely the same lesson. We have our own concentraciones now—the design systems, component libraries, and rigorous process frameworks that promise to eliminate the chaos of creative whimsy. We have our tactical drills: A/B tests, conversion funnels, growth hacking playbooks. We track our metrics with the devotion of Zubeldía’s assistant analyzing opponent set pieces. And the results, measured in quarterly growth rates and user engagement statistics, often validate the approach.

The Price of Optimization

But here is where the story becomes uncomfortable. Zubeldía’s system did not merely prioritize results; it actively suffocated the very qualities that had made Argentine football culturally significant. The gambeta was dismissed as unnecessary risk. Individual flourish was systematically coached out. As Wilson observes, Zubeldía’s philosophy was that “the match has to be won, and that’s the end of it” [1], [3], [4]. Football became not an expression of joy or identity but a zero-sum competitive exercise, stripped of aesthetic considerations.

The infamy of anti-fútbol derived not simply from its effectiveness but from its willingness to weaponize everything in service of victory. Zubeldía’s players studied opponents’ private lives to engineer psychological provocations. They mastered the “black arts” of time-wasting and referee manipulation. During the notorious 1969 Intercontinental Cup final—the “Battle of La Bombonera”—the violence became so extreme that several Estudiantes players were sentenced to prison for “disgracing a public spectacle” [1]. The system had optimized for winning so thoroughly that it had eliminated any consideration of how the victory was achieved.

This is the central paradox that contemporary design culture has yet to adequately confront. Our systems thinking, our frameworks and methodologies, our obsession with measurable outcomes, these are not value-neutral tools. They are, like Zubeldía’s tactical approach, inherently ideological. They embed certain assumptions about what matters and what can be sacrificed. And increasingly, what we sacrifice is the very dimension that creates emotional resonance and cultural significance.

Consider the contemporary obsession with “frictionless” user experiences. The term itself is revealing: friction is inherently negative, something to be optimized away. And yet friction—the slight difficulty, the unexpected detour, the moment of genuine surprise—is often where meaning emerges. As Rory Sutherland has argued, much of what makes products and experiences valuable operates at the level of the irrational, the symbolic, the psychologically resonant [5]. Strip away the friction in pursuit of efficiency, and you risk creating something technically flawless but emotionally inert.

The Metrics Trap

The problem intensifies when systems thinking intersects with measurement culture. Zubeldía’s laboratory approach was predicated on the belief that football could be decomposed into quantifiable elements: pass completion rates, territorial control, set-piece conversions. If you optimized these variables, victory would follow. The logic was impeccable. The results were undeniable. And the football was, by common consensus, unwatchable[3].

Modern design organizations have embraced a strikingly similar philosophy. Every interaction can be measured. Every design decision can be A/B tested. Every user journey can be optimized for conversion. The data, we are told, will reveal the truth. And indeed, the data often does reveal something—usually that the version with the larger call-to-action button performs 2.3% better in driving clicks.

What the data categorically cannot reveal is whether anyone actually wants to click that button, or whether the experience of doing so generates any positive emotional association with the brand. Nor can it capture the slow erosion of trust that occurs when users realize they are being relentlessly optimized. The metrics tell us that the system is working. They cannot tell us whether the system is worth operating.

This is the laboratory fallacy in its purest form: the belief that if we can measure it, we can optimize it, and if we can optimize it, we have improved it. Zubeldía measured territorial control and won matches. He could not measure—and therefore did not optimize for—the fact that his football had become so cynical that the Argentine government felt compelled to imprison his players to preserve national dignity [1].

The design equivalent manifests in products that are technically sophisticated but spiritually deadening. Everything works. Nothing delights. The buttons are perfectly sized according to Fitts’s Law. The color palette has been validated through preference testing. The copy has been optimized for reading grade level and sentiment analysis. And somehow, despite all this rigor, the experience feels like it was designed by a committee of actuaries.

When Systems Eat Creativity

There is a deeper problem embedded in systems-driven approaches: they privilege reproducibility over inspiration. Zubeldía’s genius, if we can call it that, lay in creating a framework that could turn mediocre players into a championship-winning collective. The system was the star. Individual brilliance became not merely unnecessary but actively discouraged as a source of unpredictability [1], [4].

Contemporary design organizations have embraced this logic with remarkable enthusiasm. We build design systems to ensure consistency. We create component libraries to enable “anyone” to design. We establish approval workflows to guarantee that nothing ships without proper review. The intention is admirable: to democratize design, to scale quality, to prevent the chaos of uncoordinated individual efforts.

But the effect, often unintended, is to create what might be termed “design by committee”—or worse, design by automated constraint. When every decision must fit within the system, when every component must be drawn from the approved library, when every interaction must follow the established pattern, what emerges is technically competent but creatively inert. The rough edges are sanded away. The unexpected flourishes are eliminated. What remains is professional, consistent, and profoundly boring.

This is not an argument against systems or process; both are necessary for any organization operating at scale. But it is an argument against the dangerous belief that systems alone are sufficient. Zubeldía’s training drills were valuable. His tactical innovations in pressing and spatial control were genuinely pioneering [2]. But the moment he convinced himself that football could be reduced entirely to these elements,he created something technically effective and aesthetically repellent.

The Behavioural Critique

This brings us to what might be called the behavioral critique of hyper-rationalized design. Rory Sutherland, the advertising executive and behavioral scientist, has spent much of his career demonstrating that human value perception is fundamentally irrational [5]. We prefer wine that costs more, not because it tastes better, but because the price signals quality. We find experiences more memorable when they involve mild inconvenience. We trust brands that embrace a degree of human imperfection over those that present themselves as flawlessly automated.

The implications for design are profound. If value is created not through rational optimization but through psychological resonance, then our systematic approach to eliminating irrationality is systematically eliminating value. This is not a bug in human cognition that better systems can fix; it is a fundamental feature of how meaning is constructed.

Consider the difference between Amazon’s checkout process—optimized to the point of being almost pre-conscious ritual, and the ritual of buying something from a small, well-curated shop. The former is objectively more efficient. The latter creates a memory. Amazon has optimized for conversion. The small shop has optimized for experience. These are not the same thing, and increasingly, they are inversely correlated.

Zubeldía faced an identical trade-off. He could optimize for winning, or he could optimize for the experience of watching football. He chose winning. Argentine football culture spent the next five decades wrestling with whether that was the right choice [1], [4]. The fact that the debate continues suggests that efficiency and excellence are not synonyms.

The 1958 World Cup Trauma Response

There is a psychological dimension to this story that merits examination. Zubeldía’s radical systematization was not born from pure tactical innovation; it emerged from trauma. The 1958 World Cup humiliation shattered Argentina’s confidence in la nuestra, creating what Wilson describes as a “vacuum of self-confidence” [1]. Into that vacuum stepped a totalizing ideology that promised certainty through control.

The contemporary design world exists in a similar post-traumatic state. The dot-com crash, the failures of waterfall development, the costly disasters of undisciplined creative chaos—all of these created a cultural demand for systems that could prevent future catastrophes. And so we built them: agile frameworks, design thinking methodologies, lean startup principles. Each promising that if we followed the process, we could eliminate the risk.

But systems built from fear tend toward authoritarianism. They are less concerned with enabling excellence than with preventing failure. And there is a world of difference between these goals. Preventing failure creates reliability. Enabling excellence creates possibility. The former is necessary. The latter is transformative. Confuse them at your peril.

Zubeldía prevented failure magnificently. His teams almost never lost. But they also almost never transcended. They did not inspire. They did not create moments of genuine beauty. They won, consistently and joylessly, until the watching public decided that winning on those terms was not worth the cost [1].

Generative AI: The Ultimate Anti-Fútbol

And now, just when design culture might have begun to reckon with these tensions, we face what may be Zubeldía’s apotheosis: generative AI. If anti-fútbol represented the reduction of football to optimizable variables, large language models and image generators seem to represent the complete automation of that reduction. From a corporate strategy standpoint this is the long-sought system that needs no pibe, no individual genius, no inspired deviation. Just patterns, probabilities, and the remorseless optimization of training data.

The parallels are almost too perfect. Zubeldía’s tactical repetitions find their digital heir in models trained on billions of examples to exclude uncertainty. His systematic drilling of set pieces finds its home in the fine-tuning of prompts. His reconception of players as “functional units” reaches its logical conclusion where the creative human is eliminated entirely, replaced by a statistical model of creativity itself.

The technology is remarkable, undeniably. It can generate competent designs in seconds, write passable marketing copy, produce images that would have required hours of human labour. And like Estudiantes in 1968, it wins by conventional metrics. Faster. Cheaper. More scalable. The laboratory method perfected.

But here is what should give us pause: generative AI is, by its nature, incapable of understanding why certain patterns work. It knows that certain color palettes tend to increase engagement, that certain headline structures generate clicks, that certain visual compositions test well. What it categorically cannot know is whether those patterns create meaning, whether they build trust, whether they contribute to or erode the cultural resonance that makes a brand or product genuinely valuable. It is optimizing, brilliantly and ceaselessly, for the measurable. And it cannot even conceive of what it might be missing.

This is not an argument against the technology. It is an argument that generative AI makes design strategy more essential, not less. Because if machines can now execute the tactical repetitions with perfect efficiency, the question of which tactics to deploy, what we are optimizing for, and why certain outcomes matter over others becomes the entire game. The laboratory can now run itself. But laboratories, as Zubeldía demonstrated, optimize for what they are told to optimize for. They cannot question whether winning matches is worth the moral compromise. They cannot recognize when technical success has become cultural failure.

The danger, then, is not that AI will replace designers—though it will certainly transform the profession. The danger is that we will use it to scale anti-fútbol to previously unimaginable proportions. That we will deploy it to A/B test every pixel, to optimize every word, to remove every inefficiency, every friction, every moment of human imperfection. That we will, in other words, use the most powerful tool ever created for pattern optimization to systematically eliminate precisely the elements that create meaning.

If we learned nothing from Zubeldía, we will treat generative AI as the ultimate tactical system—a way to finally, completely, exclude coincidence from creative work. We will prompt it to make things “more engaging,” “more converting,” “more efficient.” And it will deliver, magnificently. And in five years, we will look around at a digital landscape that is technically flawless and emotionally deadening, and wonder how we got here.

Finding the Balance

So where does this leave us? Are we to conclude that systems thinking is inherently destructive, that all process is creative death, that we should return to the bohemian chaos of pure intuition? That would be as naïve as Zubeldía’s opposite error: the belief that all art can be reduced to engineering.

The answer, surely, lies in understanding what systems are for—and this becomes more critical, not less, in the age of generative AI. Systems should create the conditions for creativity, not eradicate it. They should provide structure within which inspiration can flourish, not replace inspiration with mechanical execution. Zubeldía’s tactical innovations, his pressing patterns, his spatial control principles, were genuinely valuable [2]. The problem emerged only when he convinced himself that these innovations obviated the need for the gambeta, for individual brilliance, for the unpredictable joy of the pibe. Today, we risk making the same error at machine speed and global scale.

The best design organizations understand this balance. They build systems robust enough to ensure consistency but flexible enough to permit deviation. They measure outcomes rigorously while acknowledging that not everything valuable can be measured. They optimize for efficiency while preserving space for the inefficient flourish that creates emotional connection. And crucially, they understand that generative AI is a tool that amplifies whatever philosophy guides its use—whether that philosophy prioritizes pattern completion or meaning creation, optimization or resonance, efficiency or excellence.

This is harder than either pure systematization or pure chaos. It requires the discipline to build frameworks and the wisdom to know when to ignore them. It demands both the analytical rigor to decompose problems and the creative courage to resynthesize them into something that transcends mere function. It accepts that sometimes the most important outcomes—trust, delight, cultural resonance—emerge from precisely the elements that resist optimization.

When Beauty Struck Back

Argentine football eventually found its answer to Zubeldía. In 1974, César Luis Menotti became national team coach and initiated what became known as the menottista restoration: a deliberate return to la nuestra, to individual expression, to the belief that football should be beautiful as well as effective [1], [2]. Menotti’s Argentina won the 1978 World Cup playing football that was recognizably Argentine, that celebrated rather than suppressed individual talent, that understood victory and joy need not be mutually exclusive.

The tension between beauty and pragmatism continues to define Argentine football. And crucially, Argentina has won World Cups under both approaches. The country has never settled on one as definitively superior. Instead, it has learned to hold both in productive tension, recognizing that each has validity and neither has all the answers [4].

Design culture would do well to embrace a similar ambivalence. There are contexts where Zubeldía’s approach is not merely justified but necessary. When building critical infrastructure, when establishing baseline functionality, when creating the foundations upon which creativity can build—systematic rigor is essential. Generative AI will make this systematic work more efficient than ever. But when the goal is to create something that resonates, that delights, that generates the kind of emotional connection that transforms users into advocates—then the laboratory method becomes a cage, and AI becomes the lock.

The question is not whether to embrace systems or reject them, but rather when each approach is appropriate and how to prevent either from becoming totalizing. Zubeldía’s error was not that he systematized football; it was that he convinced himself systematization was sufficient. Our error, increasingly, is making the same mistake. And now we have tools that can systematize at a scale and speed that would have astonished even Zubeldía. Which makes the strategic question of what to systematize and why more essential than ever.

The Limits of Optimization

Late in his career, after the scandals and imprisonments, after Argentine football had collectively decided that anti-fútbol represented a moral dead-end, Zubeldía reportedly said: “You don’t arrive at glory through a path of roses” [1]. It was meant as a defense, a reminder that victory requires sacrifice. But it inadvertently revealed the philosophical poverty at the heart of his project. If the path to glory involves eliminating everything that makes the journey worthwhile, perhaps glory itself has been misconceived.

This is the question that contemporary design culture must confront, and with generative AI, must confront urgently. We have optimized brilliantly. Our systems work. Our metrics improve. Our processes scale. Our AI can now execute those processes with superhuman efficiency. And yet there is a growing sense—difficult to quantify, impossible to A/B test—that we are optimizing toward a local maximum, that we have mistaken efficiency for excellence, that we are winning matches while losing something more fundamental.

The laboratory method can decompose a problem, identify inefficiencies, and engineer solutions. Generative AI can do this faster than any human ever could. What it cannot do—what it is structurally incapable of doing—is generate meaning. Meaning emerges from the inefficiencies, the human imperfections, the moments of inspired deviation from the plan. It emerges when designers trust their instincts over the data, when organizations permit experiments that cannot be pre-justified by metrics, when products embrace friction because friction creates the possibility of genuine surprise. These are precisely the elements that AI, optimizing for pattern completion, will systematically remove.

Zubeldía won his trophies. His system worked, by every measurable standard. And fifty years later, Argentine football remembers him primarily as a cautionary tale, as proof that you can optimize so thoroughly for success that you forget why success was worth pursuing in the first place. The real question for contemporary design is whether we will learn that lesson before we deploy the most powerful optimization engine ever created. Or whether we will need our own Battle of La Bombonera—a digital landscape so technically perfect and so spiritually vacant that we are forced, finally, to reckon with what we have built.

The choice, as always, is not between systems and creativity, between measurement and intuition, between Zubeldía and Menotti, between AI and human designers. The choice is whether we can hold both in productive tension, whether we can build frameworks that enable rather than constrain, whether we can use our most powerful tools to create space for human flourishing rather than eliminate it. Because in the end, nobody remembers the tactical formations. They remember the gambeta. They remember the moment of individual genius that transcended the system. They remember, in other words, exactly what Zubeldía spent his career trying to eliminate—and exactly what generative AI, if deployed without strategic wisdom, will eliminate at scale.

References

[1] J. Wilson, Angels with dirty faces: The footballing history of Argentina. Orion, 2016.

[2] J. Wilson, Inverting the pyramid: The history of football tactics. PublicAffairs, 2013.

[3] O. J. Zubeldía and A. Geronazzo, Táctica y estrategia del fútbol. Jorge Álvarez Editor, 1965.

[4] J. Wilson, The barcelona legacy: Guardiola, mourinho and the fight for football’s soul. BLINK Publishing, 2018.

[5] R. Sutherland, Alchemy: The dark art and curious science of creating magic in brands, business, and life. William Morrow, 2019.

[6] Wikipedia contributors, “Disaster of sweden — Wikipedia, the free encyclopedia.” 2025. Available: https://en.wikipedia.org/wiki/Disaster_of_Sweden

Thanks for reading The Spotted Ape! Subscribe for free to receive new posts and support my work.

Throughout 2024, while South African soldiers died fighting M23 rebels in the DRC, the insurgent group’s political leadership maintained operations from a Johannesburg office building. They were raising funds, managing propaganda, and running what amounted to a shadow government. This surreal arrangement exemplified South Africa’s transformation from regional hegemon to what security analysts began calling a “continental hinterland”: the insurgent’s back office, where the messy business of modern warfare was coordinated at a comfortable distance from actual combat. Though M23’s operations have since shifted to territories under their control in eastern DRC, the pattern they established—and which other groups continue to exploit—reveals how South Africa’s institutional decay has made it essential infrastructure for transnational insurgency. All of the coordination, none of the violence. Until the violence comes home.

The Dual Reality

Few episodes better illustrated South Africa’s compromised security posture than what transpired throughout 2024. South African National Defence Force (SANDF) soldiers died in combat against the M23 rebel group in the Democratic Republic of Congo, ultimately forcing a humiliating withdrawal in February 2025 (according to an International Crisis Group assessment) [1], [2]. For much of that period, the political leadership of that very same insurgent force operated from an unremarkable Johannesburg office building, raising funds, managing propaganda, and coordinating what could only be described as a parallel state apparatus [3].

Yet M23’s playbook of diaspora mobilization, procurement networks and female-coordinated logistics wasn’t unique. It was a template other groups had already perfected. While bombs detonate in Mogadishu and the Sahel descends further into chaos, the financial plumbing, political coordination, and logistical architecture sustaining these conflicts continues to run through Cape Town, Durban, and Johannesburg. South Africa’s transformation from regional power to insurgent logistics hub represents a deliberate confluence of institutional decay, diplomatic ambiguity, and strategic incoherence. The violence happens elsewhere; Johannesburg still provides the back office.

Thanks for reading The Spotted Ape! Subscribe for free to receive new posts and support my work.

The term “continental hinterland” describes precisely this arrangement—a back office where the messy business of insurgency is conducted at a comfortable distance from actual fighting [3]. By early 2025, M23 had consolidated control over Goma, Bukavu, and other territories in North and South Kivu provinces, shifting their operational base to areas under their direct military control. Their South African presence served its purpose: establishing financial networks, procurement channels, and diaspora coordination mechanisms that could later be managed remotely. The infrastructure outlasted the physical presence.

Why Did M23 Operate Openly in Johannesburg?

Throughout 2024, the Congo River Alliance (AFC), M23’s political wing, maintained an operational hub in South Africa’s commercial capital [1], [2]. Whether the AFC was formally registered as a South African entity remains unclear from public records, but its operations were hardly clandestine—diaspora mobilization, narrative management, and according to intelligence assessments, the laundering of proceeds from mining operations in the mineral-rich Kivu provinces [3].

The cognitive dissonance bears repeating: Pretoria dispatched battalions to MONUSCO peacekeeping operations to contain an insurgency whose political leadership enjoyed freedom of operation within South Africa’s borders. Between 2023 and 2024, South African forces suffered casualties in direct combat with M23 fighters [2]. The political apparatus directing those fighters, coordinating their messaging, managing their finances, and procuring advanced military technology operated in Johannesburg, unmolested by law enforcement [3], [4].

Strategic incoherence seemed the obvious diagnosis. The more accurate interpretation: deliberate diplomatic ambiguity providing insurgent groups with a permissive operating environment. South Africa’s foreign policy establishment prided itself on “South-South solidarity,” on refusing to simply adopt Western designations of terrorist organizations, on commitment to “African solutions to African problems.” These may be noble principles. When they translated into allowing an insurgent group’s political wing to openly fundraise and procure military equipment while your own soldiers died fighting that group’s military wing, principle had curdled into something else entirely.

Women played a critical coordinating role in M23’s South African operations during 2024, particularly in managing community-level fundraising networks within Congolese diaspora populations. Female relatives of combatants maintained family connections that facilitated both recruitment and resource extraction, operating through positions of trust that made them less visible to intelligence services focused primarily on identifying male financiers and procurement specialists (based on gendered intelligence patterns documented by the Global Initiative Against Transnational Organized Crime) [5].

The Financial Architecture of Modern Insurgency

South Africa possesses, simultaneously, one of Africa’s most advanced banking sectors and one of its largest informal economies. For insurgent financiers, this is the entire point.

The Hoomer and Miller networks in Durban and Cape Town, partially disrupted by U.S. Treasury designations in 2022, operated through a clever integration of legitimate businesses and informal value transfer systems. Treasury documented transfers exceeding $69,000 to the Allied Democratic Forces in the DRC and ISIS operations across the continent (according to U.S. Treasury sanctions documentation) [6], [7]. Gold trading firms, retail operations, and import-export businesses all provided cover for what Treasury described as “many-to-one” transaction patterns, where fragmented hawala remittances coalesced into larger transfers.

The amounts might seem modest by Wall Street standards. In conflict zones, that sum buys substantial ammunition, significant bribes, and sufficient logistical capacity to sustain operations for months [7], [8]. The sophistication lies in methodology, exploiting the gap between South Africa’s formal banking oversight and its sprawling, largely unregulated informal economy.

Al-Shabaab’s fundraising apparatus demonstrates a different approach. The group relies on what it terms zakat—the extraction of obligatory Islamic charitable contributions from the Somali diaspora through religious obligation and familial coercion. There is nothing subtle about this: families in South Africa receive visits, phone calls, reminders that relatives back home might face consequences should contributions dry up. The funds flow through community networks, automotive businesses, grocery stores—sectors historically resistant to state oversight and well-suited to cash-based transactions [5], [9].

Women manage the operational backbone of these financial networks within South African diaspora communities. In Johannesburg and Cape Town, Somali women operate the small retail businesses (spaza shops, clothing stores, remittance outlets) that serve as conduits for laundering extortion proceeds. These women maintain the family connections enabling coercive fundraising, where implicit threats to relatives in Somalia ensure compliance. Traditional intelligence paradigms focus on male combatants and financiers, creating blind spots that al-Shabaab exploits systematically. The women coordinating these community-embedded networks occupy positions of trust that make them ideally placed to manage the micro-logistics of insurgent financing while remaining largely invisible to law enforcement (according to regional security assessments) [10], [11].

What makes South Africa particularly attractive is the integration of informal channels with sophisticated formal banking infrastructure. Money enters through hawala systems, launders through legitimate businesses, exits through conventional banking, all while exploiting regulatory gaps and enforcement weakness [4]. A three-card monte performed with regulatory oversight consistently watching the wrong hand.

The Hollowing of the State

None of this would be possible without what South Africans euphemistically call “state capture”—the systematic corruption and institutional erosion that characterized the Zuma administration and whose aftereffects continue to reverberate. When insurgent financiers exploit South African infrastructure, they aren’t defeating state institutions so much as operating in the vacuum left by their collapse.

The South African National Prosecuting Authority and the South African Police Service emerged from the state capture era as hollow institutions—organizationally intact yet operationally enfeebled. Critical investigative units were dismantled, experienced prosecutors driven out, intelligence capabilities allowed to atrophy. What remained was a culture of non-enforcement, where even when suspicious transactions were flagged, the capacity or will to pursue them was absent.

This institutional decay creates what might be termed “structural impunity.” Farhad Hoomer, the Durban-based businessman designated by U.S. Treasury for ISIS financing, operated for years before facing consequences (according to Treasury sanctions documentation) [7]. His network wasn’t particularly well-hidden; it simply existed in an enforcement vacuum.

The problem extends beyond individual prosecutions. The infrastructure of enforcement—interagency coordination, financial intelligence analysis, cross-border cooperation—was allowed to deteriorate during state capture and has not been rebuilt. The Financial Intelligence Centre can flag suspicious transactions; when those flags lead nowhere, when investigations stall, when prosecutions fail to materialize, technical compliance becomes theatrical. Impressive on paper, meaningless in practice.

The Grey List Paradox

South Africa’s removal from the Financial Action Task Force grey list in October 2025 was hailed by government as vindication of counter-financing reforms. From a technical standpoint, the reforms were substantial: enhanced customer due diligence, improved beneficial ownership transparency, strengthened reporting requirements. The kind of measures that look impressive in FATF evaluation reports [12].

Technical compliance can coexist with operational ineffectiveness. FATF assessments evaluate whether regulatory architecture exists, not whether it actually disrupts illicit finance. South Africa built the architecture. Whether it functions as intended remains a different question entirely.

The grey list removal may actually worsen the problem. It reduces international scrutiny precisely when that scrutiny might be most needed. It signals that the money laundering problem has been addressed when, in reality, only the regulatory framework has been updated. The networks that sustained ISIS financing before October 2025 didn’t dissolve with South Africa’s delisting; they continued operating beneath a more polished regulatory facade.

The emphasis on formal banking sector compliance misses the point when much problematic activity occurs in the informal economy. Enhanced bank supervision becomes irrelevant if money never touches a bank. Hawala networks, cash-based businesses, community-embedded extortion—none of this falls neatly within FATF’s evaluation criteria, which are designed for formal financial systems.

Even more troubling are sectors that lubricate insurgent logistics while evading quantitative capture entirely. Illegal mining operations (the so-called zama-zamas) and cigarette smuggling networks provide both revenue streams and operational cover for insurgent financing [12], [13]. These informal economies exist in a regulatory twilight zone, too dispersed to monitor effectively, too embedded in local communities to disrupt without significant political cost.

The Emerging Threat of Domestic Blowback

South Africa’s role remains primarily logistical—coordination without combat, infrastructure without carnage. This arrangement contains the seeds of its own transformation. Hinterlands have a nasty habit of becoming frontlines.

The dynamics are straightforward: When insurgent groups establish deep logistical roots in a territory, they develop local networks, recruit local operatives, create dependencies. Over time, these groups acquire the capability and potentially the motivation to conduct operations within their host territory. ISIS affiliates that currently use South Africa for fundraising and coordination could, if circumstances warrant, pivot to using it as an operational theater.

The deepening organizational structure—the recruitment of families, the embedding in community networks, the development of female-managed logistical architectures—indicates these groups aren’t merely passing through. They’re putting down roots. The more embedded they become, the more vulnerable South Africa becomes to domestic blowback that transforms rear bases into frontline targets.

Historical precedents exist, all pointing toward escalation. London served as a rear base for various Middle Eastern militant groups throughout the 1990s until it didn’t. The permissive environment that made it attractive as a logistical hub eventually made it attractive as a target. The 7/7 bombings were conducted by British citizens radicalized within networks initially established for external operations.

South Africa need not replicate London’s experience, but the risk factors are alarmingly similar: deep diaspora networks, institutional weakness, enforcement gaps, and a political establishment that mistakes non-alignment for immunity. The longer these networks operate, the more embedded they become, and the more difficult they are to disrupt without triggering precisely the kind of domestic blowback that officials currently believe they’re avoiding.

What Happens Next?

South Africa finds itself in a paradoxical position: removed from the FATF grey list while functioning as a critical node in continental insurgent infrastructure; having hosted M23’s political operations while sending soldiers to die fighting them; possessing sophisticated anti-money laundering regulations while insurgent financing networks operate with relative impunity.

The fundamental problem is political will masquerading as technical capacity. South Africa’s government has demonstrated it can build regulatory architecture when international pressure demands it. What it has not demonstrated is willingness to actually use that architecture to disrupt the networks that matter. Designating a few financiers to satisfy American demands is one thing; systematically dismantling hawala networks, prosecuting community leaders engaged in extortion, and proscribing insurgent political fronts is quite another.

Such systematic disruption would require confronting an uncomfortable reality: South-South solidarity and non-alignment have practical limits. There is no particular virtue in providing sanctuary to groups engaged in mass violence, regardless of whether Western governments have designated them as terrorists. It would require rebuilding prosecutorial and intelligence capacity gutted during state capture. It would require developing intelligence frameworks that can actually identify and disrupt the female-managed logistics networks that current approaches systematically overlook—frameworks that recognize women’s roles in al-Shabaab’s zakat collection, M23’s diaspora coordination, and ISIS affiliate financial management as central to operational success rather than peripheral support functions.

Above all, it would require political courage.

Concrete policy measures that should be implemented immediately:

South Africa’s Financial Intelligence Centre requires two immediate interventions: First, forensic analysis of M23’s historical financial flows to identify ongoing nodes—the businesses, intermediaries, and community structures that don’t dissolve when clients relocate but simply find new ones. Second, emergency funding for a dedicated informal economy monitoring unit focused on hawala networks and cash-intensive sectors like spaza shops, automotive businesses, and gold trading that current formal banking compliance frameworks cannot touch. Al-Shabaab, ISIS affiliates, and other groups continue to exploit precisely these gaps.

Law enforcement must develop gender-sensitive intelligence training programs that recognize female-managed logistical networks as primary targets rather than peripheral concerns. The women coordinating community fundraising, managing small retail businesses, and maintaining diaspora connections aren’t supporting roles in insurgent infrastructure—they are the infrastructure. M23’s 2024 operations demonstrated this; al-Shabaab’s ongoing networks prove it daily.

Until such measures materialize, South Africa will continue its role as continental hinterland. M23 may have relocated to territories under their control, but the financial flows that sustained their 2024 operations continue through other channels and other groups. Al-Shabaab’s extortion networks expand, ISIS affiliate financing persists, and the female-managed logistical architectures deepen. The infrastructure M23 helped establish remains available, adaptable, and increasingly sophisticated.

The question isn’t whether this arrangement is sustainable—it manifestly isn’t—but when and how it collapses. A gradual awakening to the threat, accompanied by genuine institutional revitalization, remains possible. So does an alternative: waiting for a domestic atrocity to concentrate minds.

If there’s one lesson from the last two decades of counter-terrorism, it’s this: by the time the violence reaches home, it’s already far too late to prevent it. The insurgent’s back office doesn’t remain a back office forever. Eventually, inevitably, it becomes the front.

South Africa still has time to avert that future. Time, like institutional capacity and political will, is not an infinite resource. The meter is running.

References

[1] Britannica, “March 23 movement: Leaders, funding, and AFC.” 2025. Available: https://www.britannica.com/topic/March-23-Movement

[2] International Crisis Group, “The M23 offensive: Elusive peace in the great lakes,” Africa Report 320, 2025.

[3] Critical Threats Project, “M23’s state building project: Africa file special edition,” Special Report, 2024.

[4] South African Reserve Bank Prudential Authority, “Money laundering, terrorist financing and proliferation financing sector risk assessment for the south african banking sector,” Sector Risk Assessment, 2025.

[5] Global Initiative Against Transnational Organized Crime, “The terror-crime nexus in south africa: Informal retail and transport extortion,” Research Report, 2025.

[6] The Soufan Center, “IntelBrief: South Africa has emerged as a financial hub for ISIS in Africa.” 2023. Available: https://thesoufancenter.org/intelbrief-2023-march-21

[7] A. Tobin, “U.S. Treasury designates four ISIS financiers in South Africa,” FDD’s Long War Journal, 2022.

[8] E. Sanderson, “The islamic state in South Africa: A 6-month outlook,” Grey Dynamics, 2022.

[9] E. O. S. Odhiambo, E. W. Njoroge, and M. M. Hafez, “Al-Shabaab’s financial fortitude: The link between funding strategies and organizational resiliency,” Open Journal of Political Science, vol. 15, no. 2, 2025.

[10] M. Ewi, “Boko Haram’s South African connections: What do we know?” ISS Today, 2022, Available: https://issafrica.org/iss-today/boko-harams-south-african-connections-what-do-we-know

[11] H. Matfess, Women and the war on Boko Haram: Wives, weapons, and witnesses. Zed Books, 2017.

[12] N. Stewart, “Policy alert: Congress must act to keep South Africa on FATF grey list,” Foundation for Defense of Democracies, Policy Alert, 2025.

[13] S. Haysom, “In South Africa, illicit cigarettes are a smoking gun on corruption,” New Atlanticist, 2019.

Thanks for reading The Spotted Ape! Subscribe for free to receive new posts and support my work.

When President Trump declared last week that acquiring Greenland represents “a national security priority” and his press secretary confirmed that “utilising the U.S. Military is always an option” [1], the pronouncement generated predictable outrage. Danish officials pushed back forcefully [2], whilst NATO allies issued stern statements about sovereignty [3]. At the same time in Svalbard, where China’s Yellow River Research Station has stood since 2004, Norway ordered the removal of symbolic stone lions from the facility’s entrance—a telling indication of how even ostensibly scientific Arctic presences have become sites of geopolitical contestation [4].

Greenland and Svalbard are critical military nodes; the S2S nexus enables hybrid operations, legal gaps are exploited strategically, and multi-domain control links communications, missile defense, and intelligence collection.

Yet, the motive might have less to do with minerals than with a technical reality: physics. The Arctic has become strategically indispensable because of the orbital mechanics that govern satellites. Whilst Trump noted that “Greenland is covered with Russian and Chinese ships” [5], the real competition is in data streams connecting the seafloor to orbit.

The traditional concepts of territorial sovereignty prove spectacularly ill-suited to contemporary strategic requirements. These contests do not unfold at the level of flags or borders, but in the ambiguous spaces where infrastructure, science, and strategy overlap—spaces in which presence itself has become power.

Consider a reconnaissance satellite. Polar orbits provide global coverage but converge at the poles. A ground station at 40°N might only contact a satellite three times daily [6]. In contrast, Greenland’s Pituffik Space Base at 76°N enables “14 of 14 capability”—near-continuous data collection [7].

Geography asserts itself where technology cannot. The Center for Strategic and International Studies (CSIS) observes that “orbital access scarcity” is determined by location, not innovation [8]. States controlling Arctic territory control the bottlenecks of space-based intelligence.

Arctic ground stations are integrated with subsea fibre-optic cables. This creates a “subsurface-to-space nexus”: severing cables degrades orbital capabilities without touching the satellites themselves [9].

> Strategic Note: Attacks on cables preserve plausible deniability. One drags an anchor across a route, attributes it to error, and accomplishes strategic degradation below the threshold of war [8].

Recent incidents involving the Chinese-flagged Yi Peng 3 and the severance of Svalbard’s cables highlight this vulnerability [10], [11]. The Greenland-Iceland-United Kingdom (GIUK) Gap serves as both a submarine transit point and a cable concentration zone [12].

In addition, the legal framework governing the Arctic is outdated. The 1967 Outer Space Treaty and 1982 United Nations Convention on the Law of the Sea (UNCLOS) did not anticipate ground station positioning or dual-use research facilities [13], [14]. This “legal lacuna” has allowed China’s Military-Civil Fusion doctrine to flourish in scientific installations [15].
In conclusion, geography has reasserted itself. Future military power will depend on the infrastructure linking Earth to orbit [16], [17]. Control of the Arctic infrastructure (i.e. the ground stations and cables) implies controlling the connective tissue of modern power [18].

References

[1] K. Leavitt, “White house statement on Greenland,” NBC News, 2026.

[2] M. Frederiksen, “Statement on U.S. Greenland threats,” CNN, 2026.

[3] European Leaders, “Joint statement on Greenland,” Al Jazeera, 2026.

[4] T. Nilsen, “‘This decision is not likely to be appreciated by Beijing’: Two stone lions outside the door to the Yellow River Station in Ny-ålesund at Svalbard could cause turmoil for Norway–China science cooperation on Arctic.” [Online]. Available: https://www.thebarentsobserver.com/news/this-decision-is-not-likely-to-be-appreciated-by-beijing/428405

[5] D. Trump, “Remarks aboard Air Force One,” CNN, 2026.

[6] G. Falco, N. Boschetti, and I. Nikas, “Undercover infrastructure: Dual-use Arctic satellite ground stations,” CIGI Papers, no. 291, 2024.

[7] M. Lynch, “Space and defense in the high north,” Space and Defense, vol. 16, 2025.

[8] Center for Strategic and International Studies, “Addressing arctic vulnerabilities,” 2025. Available: https://www.csis.org/analysis/addressing-arctic-vulnerabilities

[9] G. Falco, “The arctic: A new high-ground battleground for space security,” CIGI Policy Brief, no. 184, 2024.

[10] A. Lott, “Conventional jurisdictional approaches to submarine cables,” International and Comparative Law Quarterly, vol. 74, no. S1, 2025.

[11] SIPRI, “A legislative route to combat sabotage of undersea cables.” 2025. Available: https://www.sipri.org/commentary/topical-backgrounder/2025/legislative-route-combat-sabotage-undersea-cables

[12] T. Hughes, “Closing the gap between Europe and North America,” High North News, 2024.

[13] UNOOSA, “Outer space treaty.” 1967. Available: https://www.unoosa.org/oosa/en/ourwork/spacelaw/treaties/outerspacetreaty.html

[14] A. L. Kahraman, “China’s digital footprint in the Arctic,” The Arctic Institute China Series 2025, 2025.

[15] Center for Strategic and International Studies, “Frozen frontiers: China’s great power ambitions,” 2024. Available: https://features.csis.org/hiddenreach/china-polar-research-facility/

[16] U.S. Department of Defense, “2024 department of defense arctic strategy,” 2024. Available: https://media.defense.gov/2024/Jul/22/2003507396/-1/-1/1/2024-DOD-ARCTIC-STRATEGY.PDF

[17] FPRI, “Protecting the north American Arctic and beyond,” 2025.

[18] K. Frazier, “Dual-use dilemma: Maritime infrastructure and the Arctic,” Ted Stevens Center for Arctic Security Studies, 2025.

Consider the baobab (Adansonia digitata). A tree that looks as if it has been planted upside-down. A great, gnarled trunk surmounted by a tangle of branches that resemble roots clawing at the African sky. It is preposterous and magnificent. It has, for millennia, been quietly conducting a masterclass in how to live, how to know, and how to belong.

The baobab is not merely a tree. It is a metaphor for relational knowledge sytems. The Western epistemic tradition, if we can call it that, resembles a taproot system. A single root drills straight down in search of a foundational truth. It is an approach that demands uniformity, thrives on binaries (Good/Evil, Us/Them, Centre/Periphery), and has an unlovely habit of trying to colonise everything it encounters, including our ways of knowing [2]. This is the epistemology of the grid, an extractive industrial monoculture. It is linear, hierarchical, and remarkably efficient at turning living systems into resources.

The baobab, and the African epistemologies it embodies, proposes something altogether more entangled. Its root system spreads horizontally, a rhizomatic tangle without a single centre. This, as scholars like Édouard Glissant have argued, is the model for Relation. He proposes that identity and wisdom emerges from chaos-monde - the messy, reciprocal networks of contact in life [2]. It is a wisdom of connection, not conquest. As an old Akan and Ewe proverb has it:

“Nunya, adidoe, asi metunee o”
“Wisdom is like a baobab tree; no one individual can grasp it” [2].

One cannot embrace the literal tree. Its girth can exceed eleven meters, and one cannot monopolise the wisdom it represents. It is collectively held, communally sustained.

This relational wisdom is unmistakable among the Hadza people of northern Tanzania. As one of the world’s last remaining hunter-gatherer communities, they offer a rare, continuous record of our human patterns of life. Their existence is not alongside the baobab, but within its ecological and spiritual web. The tree is their supermarket, pharmacy, hardware store, water reservoir, and cathedral. Its hollow trunks trap monsoon rains, creating natural cisterns (kesati) that the Hadza access with peg ladders [3]. Its fruit, naturally dehydrated on the branch, are pulped into a vitamin C-rich “milk”. The bark becomes rope and bowstring. The pods become cups and carriers. Bees nest in its heights, and a remarkable bird, the honeyguide, leads the Hadza to the hives in a feat of interspecies cooperation.

This deep, symbiotic adaptation treats the tree not as a resource to be extracted, but as a relation to be sustained. The Hadza even give birth within the baobab’s hollow trunks, remaining there until the umbilical cord falls away: a poignant literalisation of being born into a web of life [3]. The tree features in their creation myths as the axis along which ancestors descended from the sky. They engage with the spiritual realm associated with it, but do not worship the tree itself [3].

Our modern taproot-world has also “discovered” the baobab. It is marketed as a superfood, its pulp packed with prebiotic fibre. Ironically, this very fibre is likely a key reason Hadza gut microbiomes are the most diverse on Earth, a bastion of health that industrial diets have eroded [3]. The extractive taproot mentality now competes with the Hadza for the fruit, threatening to sever the very relational web that sustains both the people and, arguably, the trees themselves. As researchers bluntly note, “The fate of the Hadza will be the fate of their baobab trees” [3].

Here lies the dilemma. The baobab’s symbolic power is so potent it can be weaponised for opposite ends. In some political discourse, it is cast as a monstrous, obstructive state that is “damned hard to uproot” [4]. Yet in its cosmic role, it is the great provider, the tree of life.

The rhizomatic, relational wisdom embodied by the baobab and lived by the Hadza offers a corrective to our age of catastrophic bipolar thinking. It suggests that knowing is not about domination and delineation, but about connection and context. That belonging is not about purity of origin, but about the generative tangle of relationships. That a tree, a people, or an idea, can be singular in presence, yet at the same time defiantly plural in meaning.

In our monoculture of thought, obsessed with singular origins and “one best way,” the upside-down tree stands as a right-side-up challenge. Its wisdom is not owned. It is shared. It is not declared but enacted. It asks, quietly but persistently, whether we are brave enough to let go of the taproot and learn, finally, how to embrace the tangled mesh of life.

References

[1] F. Reus, “Baobab and elephant, tanzania.” Wikimedia Commons, Creative Commons Attribution-ShareAlike 2.0, 2004. Available: https://commons.wikimedia.org/wiki/File:Baobab_and_elephant,_Tanzania.jpg

[2] K. N. Ngwa, A. Cisse Niang, and A. Pressley, Eds., Life under the baobab tree: Africana studies and religion in a transitional age. Fordham University Press, 2023.

[3] J. Rashford, Baobab: The Hadza of Tanzania and the baobab as humanity’s tree of life. Springer, 2023.

[4] N. Thompson and S. Thompson, The baobab and the mango tree: Lessons about development: African and Asian contrasts. Zed Books, 2000.

“For to everyone who has, more will be given.” — Matthew 25:29

When Pierre Azoulay analysed citation patterns among life scientists in 2014, he discovered something peculiar [1]. Papers by newly minted Howard Hughes Medical Institute investigators saw their citations jump by 12% overnight - not because the research improved, but because the letterhead changed. More tellingly, the boost nearly doubled for work published in lower-tier journals, where quality signals were murkiest. The finding vindicated a hypothesis sociologist Robert Merton had floated four decades ago [2]: in domains clouded by uncertainty, prestige doesn’t just predict success, it manufactures it.

This is the Matthew Effect in its purest form, named for the Gospel passage where “to those who have, more will be given.” Yet Merton’s 1968 essay in Science, and his richer 1988 follow-up in Isis [3], revealed something more profound than biblical coincidence. The Matthew Effect isn’t a bug in meritocracy’s code; it’s the operating system. And status branding (i.e. the deliberate cultivation of prestige as identity capital) is the mechanism that keeps the gears turning, converting minor early advantages into self-reinforcing dominance with the efficiency of compound interest.

The Catalyst: When Affiliation Becomes Achievement

Consider two lawyers, equally talented, graduating in the same cohort. One carries a Harvard credential; the other, a respectable regional university. By age 35, the Harvard alumnus will likely out-earn her peer by 30%, not because she litigates better cases, but because clients and partners encode her institutional affiliation as a proxy for competence [4]. A fundamental substitution has taken place: her worth is no longer assessed by examining her own capabilities, but rather by invoking the prestige of the institution that credentialed her. Her value is defined through identity and affiliation rather than intrinsic merit.

This is status branding as catalyst. Like a chemical reaction requiring an initial spark, cumulative advantage needs a starting asymmetry. Merton termed it an “institutional location” [3]. Top departments accelerate recognition, while elite schools open boardrooms. The mechanism is behavioural. Audiences, facing Knightian uncertainty about quality, resort to prestige hierarchies as cognitive shortcuts [5]. We cannot assess the lawyer’s true skill, so we default to the halo of her alma mater, much as 19th-century consumers judged craftsmanship by guild marks rather than tactile inspection.

The process compounds in a self-amplifying loop. [3]. Brands like Ferrari and Apple have mastered this. By limiting production and controlling supply, they ensure each product becomes a symbol of prestige and extraordinary craftsmanship, transforming ordinary items into coveted treasures through perceived scarcity [4]. Owning a Birkin bag or Harvard MBA confers status merely by association, initiating what Merton observed in science. Early fame speeds idea diffusion via citations, now scaled across consumer capitalism. The credential becomes the product; the product, a status symbol; the symbol, a self-fulfilling prophecy.

The Shortcut: Uncertainty as Amplifier

Why does this work so reliably? Because quality, in most domains, is illegible. Merton called this the “Other Uncertainty Principle”: absent direct evidence of merit, we triangulate via prestige. Venture capitalists cannot predict which startup will succeed, so they favour Y Combinator alumni. Publishers cannot guarantee bestsellers, so they overpay authors with Penguin Random House pedigrees. Art collectors cannot objectively rank contemporary painters, so they defer to Sotheby’s provenance [4].

Azoulay’s team found this amplification effect most pronounced when uncertainty about quality was highest. Publishing in lower-profile journals or exploring novel research areas provided more reason to doubt the science before the prize [1]. Status serves as social proof under fog, much as Cialdini described in Influence [6]. When unsure, we copy the crowd, or, more precisely, we copy whoever the crowd has already crowned. The citation premium for high-status scientists isn’t modest. It’s a structural phenomenon, embedded in how knowledge diffuses through networks where fame begets fame in a self-accelerating cycle[3].

The digital economy has supercharged this dynamic. Platforms reduce complex relationships to visual displays of quantity (i.e. followers, likes, citations). These displays serve as currency, often enshrining the Matthew Effect in their code base by algorithmically pushing content that is already gaining momentum [7]. Google Scholar’s ranking algorithm, critics note, weights citation counts heavily, ensuring highly cited papers appear in top positions and gain ever more citations, while new work struggles to surface. LinkedIn’s verification badges and Instagram’s blue ticks are prestige capture in microcosm. The algorithms reward the already-visible, turning personal branding into a digital Matthew trap [4].

The Flywheel: Self-Fulfilling Prophecy of Prestige

Once triggered, the Matthew Effect becomes circular. A product is deemed meritorious because it carries the logo of an anointed producer. The producer gains higher status for making such a “good product” [4]. This is not cynicism but economic rationality under incomplete information. Prestige pricing strategies rest on the belief that certain customer segments will pay more. To earn this premium, brands build narratives around legacy and unique value. They showcase exceptional quality through refined design and superior craftsmanship, and they build exclusivity through limited availability and scarcity.

Hermès exemplifies this alchemy [8], [9]. The Birkin bag’s mystique rests less on its exquisite leather than on its controlled scarcity and celebrity associations. From 1980 to 2015, Birkins appreciated at approximately 14% compound annual growth rate, outperforming fine jewelry and even the S&P 500 [9]. Hermès doesn’t just make handbags; it curates an experience of aspiration through invitation-only sales systems and waiting lists spanning over six years [9]. Buyers aren’t purchasing utility; they’re acquiring membership in a status hierarchy [4], much as Harvard’s endowment snowballs not from investment genius alone but from the reputational compounding of alumni networks. The credential or product becomes symbolic property, asserting dominance in a zero-sum game for recognition.

Merton anticipated this in his analysis of intellectual property [3]. Scientists exchange knowledge freely for “reputational pellets” (i.e. citations) that aggregate into career capital. But this communism of ideas, meant to accelerate progress, breeds inequality. Priority disputes arise because credit is finite. Newton’s “standing on ye shoulders of Giants” was less an act of homage than a claim of intellectual property. In consumer markets, the same logic plays out. Brands don’t compete on quality alone but on the symbolic weight of their names, converting belief into outcome through feedback loops that solidify top dominance.

The Inequality Machine

The consequences are stark. Measures of academic success follow extremely right-skewed distributions. A minority of all academics produce the most research output and attract the most citations, with a 2018 study of science funding in the Netherlands finding that winners just above the funding threshold accumulated more than twice as much funding during the subsequent eight years as non-winners. Merton documented this in 1988 [3]: 28% of US federal research funds flowed to just 10 universities; 0.3% of articles garnered over 100 citations while 58% limped along with one.

DiPrete and Eirich’s 2006 review in Annual Review of Sociology [10] framed this as a mechanism for inequality, not just an outcome. Status branding initiates the process. The funding, collaborators and publication slots concentrate among the anointed. The gaps widen longitudinally. Eminent scientists accrue larger increments of peer recognition for equivalent contributions, not because they work harder but because their reputations insulate them from scrutiny and amplify their visibility [3]. This is the Reverse Matthew Effect: high status protects against negatives. When retractions occur, Azoulay found [1], less eminent coauthors suffer greater reputational damage. Blame deflects upward to prestige, downward to obscurity.

The result is talent wastage. Merton worried about late bloomers. These “mutants” were found to drop out from elite milieus due to a preference for early bloomers [3]. Stanovich’s 1986 work on reading literacy found similar patterns [11]: early status gaps in primary school compound lifelong, disadvantaging children who bloom later cognitively. Social class exacerbates this. Privileged late bloomers persist via networks; working-class prodigies drop out when initial advantages fade. The Matthew Effect, left unchecked, doesn’t just reward merit unequally. It obscures merit entirely, mistaking pedigree for performance [4].

The AI Acceleration

As artificial intelligence (AI) floods information markets, status heuristics grow more potent, not less. Faced with abundance, humans double down on reputation scarcity. We cannot parse a million research papers, so we defer to Nature’s anointing. We cannot vet a thousand fintech startups, so we follow Sequoia’s portfolio. AI was supposed to democratise access. Instead, it has amplified the halo of anointed creators. OpenAI’s early backers enjoy disproportionate valuations not from superior products but from association with the buzziest brand in tech. The algorithm, supposedly neutral, encodes yesterday’s prestige into tomorrow’s opportunities [4].

This phenomenon has birthed masstige marketing [12], [13]. Masstige targets the middle class by combining luxury brand values with mass-market access, promising consumers superior experience, engagement, and status through expensive or premium products marketed to many. Apple captured 46% of global smartphone revenue in 2024 while holding just 19% of market share [14] — prestige pricing at planetary scale. Nike’s Air Jordans function as collectibles, their value inflating through timed releases and artificial scarcity. These aren’t products; they’re status signalling devices, repackaging exclusivity for mass consumption [13].

The danger is expropriation without credit - Merton’s nightmare scenario [3]. When ideas circulate freely but attribution collapses under algorithmic curation, late-stage creators reap rewards while originators languish in obscurity. LinkedIn’s blue-tick economy exemplifies this: verified influencers amplify each other’s content, creating closed loops of visibility while unverified voices shout into the void. The prestige economy becomes a zero-sum tournament where network effects entrench incumbents, and new entrants need exponentially more talent to break through [4].

The Randomness Solution: Betting on Unknowns

Can we escape this trap? The conventional answer of diversity initiatives, blind review, and credential-obscuring misses the deeper problem. These interventions still optimise for past performance, retrofitting yesterday’s patterns onto tomorrow’s talent. They’re actuarial thinking dressed in progressive language, measuring historical correlations and rebalancing portfolios of known quantities [4].

Clayton Christensen grasped the core dilemma [15]. When managers wait until data is clear, the game is over. Actuarial models excel at pricing existing risks (e.g. mortality tables, credit defaults, insurance claims) but catastrophically fail at predicting disruption. They look backward from the iceberg’s impact, calculating damage with precision, then steer straight into it. The future belongs not to those who optimise for the known past, but to those who inject sufficient randomness into selection processes. Merton called them the “countervailing forces” against cumulative advantage [3].

This requires magnanimity toward failure, a tolerance for backing unknowns. Funding lotteries, as practiced by New Zealand’s Health Research Council and increasingly advocated by scientists can randomise early advantages. They’re not meritocratic in the credentialist sense; they’re anti-Matthean. By selecting proposals that clear a quality threshold assisted by a lottery process, rather than expert ranking, they eliminate the institutional halo. Failure becomes permissible, even necessary, because the system no longer pretends omniscience about potential [4].

Contrast this with planned diversity hires, which often backfire by encoding new status markers. Hiring for “diverse backgrounds” while demanding Ivy League credentials simply reshuffles prestige hierarchies without disrupting them. True disruption means choosing candidates with no legible pedigree, no traceable prestige DNA. These may include immigrants from unranked institutions, late bloomers without publications, autodidacts without portfolios. It means betting on potential futures rather than auditing actuarial pasts [15].

The iceberg metaphor applies here too. Organisations fixated on visible credentials (i.e., the proverbial tip above water) miss the submerged mass of latent talent. Merton’s late bloomers [2] attrite precisely because systems optimise for early-stage visibility. But early achievement is often privilege: access to resources, networks, and institutional location that accelerate recognition. The child who publishes at 22 had advantages the night-shift worker studying at 32 did not. Randomness corrects for this by refusing to privilege legibility itself[4].

The Prestige Paradox

Status branding is both currency and cage. For the rich, it’s protective, insulating against failure. For the rest, it’s punitive, deflecting blame downward. Merton likened science’s free exchange of knowledge to an uncommon good. It grows through use rather than depleting like pastures [3]. But this good, secured by symbolic property, breeds inequality as surely as enclosures did. The Matthew Effect isn’t a moral failing; it’s an emergent property of how humans navigate uncertainty under resource constraints [4].

We cannot abolish prestige. Reputational hierarchies are cognitive necessities in a world too complex to audit fully. But we can refuse to let them calcify into destiny. The answer isn’t diversifying the anointed. It’s choosing the unknown. Rotate collaborations so junior researchers work with senior stars to diffuse prestige via proximity rather than inheritance. Strip author names from grant proposals during review. Limit citation counts visible to editors to force engagement with ideas rather than CVs. Weight variance over consistency. Reward wild swings over steady incrementalism, because disruption requires tolerating failure [15].

The true Matthew miracle, then, isn’t spotting talent before the halo descends. It’s backing candidates before legibility emerges. As Christensen observed [15], theory beats data for predicting the future. Gravity predicts you’ll fall from a cliff without requiring historical evidence. The theory here is simple: prestige compounds existing advantages, so inject randomness to disrupt compounding. Choose unknowns as part of a cohort. Permit failure. Bet on potential, not pedigree.

In a prestige economy rigged to reward yesterday’s winners, the most subversive act isn’t refining selection. It’s admitting we cannot predict genius and designing for that humility. The ship hits the iceberg because actuaries optimise for the past. The future belongs to leaders willing to steer into fog, knowing some will fail but others will discover new worlds. Magnanimity toward failure isn’t soft; it’s the only escape from cumulative advantage’s iron law [4].

References

[1] P. Azoulay, T. Stuart, and Y. Wang, “Matthew: Effect or fable?” Management Science, vol. 60, no. 1, pp. 92–109, 2014.

[2] R. K. Merton, “The Matthew Effect in science,” Science, vol. 159, no. 3810, pp. 56–63, 1968.

[3] R. K. Merton, “The Matthew Effect in science, II: Cumulative advantage and the symbolism of intellectual property,” Isis, vol. 79, no. 4, pp. 606–623, 1988.

[4] T. E. Stuart, Anointed: The extraordinary effects of social status in a winner-take-most world. New York: Simon & Schuster, 2025.

[5] D. Kahneman, Thinking, fast and slow. New York: Farrar, Straus; Giroux, 2011.

[6] R. B. Cialdini, Influence: The psychology of persuasion, Revised. New York: Harper Business, 2006.

[7] M. Perc, “The Matthew Effect in empirical data,” Journal of the Royal Society Interface, vol. 11, no. 98, p. 20140378, 2014.

[8] The Fashion Law, “How Hermès turned the Birkin bag into a multi-billion dollar business.” [Online]. Available: https://www.thefashionlaw.com/how-hermes-turned-the-birkin-bag-into-a-multi-billion-dollar-business/

[9] Quartr, “The birkin bag: The story behind the world’s most desired handbag.” [Online]. Available: https://quartr.com/insights/company-research/the-birkin-bag-the-story-behind-the-worlds-most-desired-handbag

[10] T. A. DiPrete and G. M. Eirich, “Cumulative advantage as a mechanism for inequality: A review of theoretical and empirical developments,” Annual Review of Sociology, vol. 32, pp. 271–297, 2006.

[11] K. E. Stanovich, “Matthew Effects in reading: Some consequences of individual differences in the acquisition of literacy,” Reading Research Quarterly, vol. 21, no. 4, pp. 360–407, 1986.

[12] M. J. Silverstein and N. Fiske, “Luxury for the masses,” Harvard Business Review, vol. 81, no. 4, pp. 48–57, 2003.

[13] A. Kumar, J. Paul, and A. B. Unnithan, “’Masstige’ marketing: A review, synthesis and research agenda,” Journal of Business Research, vol. 113, pp. 384–398, 2020.

[14] Counterpoint Research, “Apple captured 46 percent of the global smartphone revenue share for 2024.” [Online]. Available: https://wccftech.com/global-smartphone-revenue-2024-apple-took-the-majority-share-with-46-percent/

[15] C. M. Christensen, The innovator’s dilemma: When new technologies cause great firms to fail. Boston, MA: Harvard Business School Press, 1997.

Does your team have a Spotted Ape? A brightly coloured B-24 Liberator from 1943 offers a lesson for the AI age: creative mentors who can bring order to chaos with a light touch.

The Spotted Ape - also called the Flying Spotted Ass Ape - was a battered bomber splashed with polka dots, circling East Anglia like a circus act that refused to land. Too weary for combat but too useful to discard, it had one task - to gather hundreds of bombers into formation before they vanished into the fog.

The skies over England were crowded and confused. Green crews lifted off from dozens of fields, often colliding before reaching altitude [1]. On 1 August 1943, during the raid on Ploiești, disarray in assembly and navigation cost 53 out of 177 aircraft and 310 lives [2]. The answer was absurd and brilliant: turn one old plane into a beacon [1]. Painted in clownish colours and fitted with flares, it would climb first, circle a fixed point, and signal until its flock formed [1]. When the bombers were ready, it dipped beneath the lead ship and went home [1]. The crews called it the “Judas Goat” - the guide that led others to battle but never joined the slaughter [1].

Lessons from a polka-dotted bomber

These garish aircraft became standard in late 1943 and 1944 as the Eighth Air Force expanded from a few hundred bombers to armadas of more than a thousand [1], [3]. During Big Week in February 1944, over a thousand bombers struck the German aircraft industry [4]. Similar numbers flew in the weeks before and after D-Day, hitting multiple targets [3]. By early 1945, raids on Berlin sent 1,400 bombers and nearly a thousand fighters into the air [3].

With such vast formations, order mattered more than firepower [1]. The assembly ships were not weapons but anchors. The 458th Bombardment Group at RAF Horsham St Faith had one of the most famous: the Flying Spotted Ass Ape, also known as the Spotted Ape [1]. For nearly a year, piloted by experienced crews, it guided every major mission without firing a shot [1]. The genius lay not in its technology but its psychology — visibility, credibility, and timing.

The same principles apply to modern creative teams, especially those navigating the turbulence of generative AI. Technology accelerates ideas but amplifies noise. Brilliant individuals chase their own trajectories. Deadlines loom. Meetings collide. Everyone is flying, but no one is forming. What the Spotted Ape offered in the air, creative teams now need on the ground: visible cues, calm example, practical wisdom, and the discipline to withdraw once the group finds its rhythm.

Example

Assembly ships did not command; they demonstrated. When one aircraft steadied its course, others followed. Leadership through example remains stronger than instruction [5]. When a senior creative admits uncertainty or pauses to recalibrate, it grants permission for others to do the same. Formation follows behaviour, not memos.

Experience

The Spotted Ape was flown by veterans who knew turbulence, fuel strain and fear — the feel of the craft itself. Michael Polanyi called this “tacit knowledge,” the unspoken skill that resists codification [6]. An “indwelling” of routines and shared heuristics [6]. Modern leaders often hover above the work they manage. The best still know how to fly.

Exit

Once the bombers held formation, the Spotted Ape turned home [1]. Leadership that withdraws at the right moment leaves competence behind. Leadership that lingers breeds dependence. In many AI-driven teams, constant oversight creates learned helplessness. True guidance ends in self-correction.

Deploying a modern Spotted Ape

Launching a modern Spotted Ape is about people, not platforms. Start complex projects with a seasoned guide for the first sprint — someone who reads personalities, spots friction early, and sets small rituals of alignment before the pressure sets in. Their role is to kindle autonomy, not control it. Once the rhythm holds, they step aside. Shared experiences humanise complexity. Mistakes become collective learning rather than individual blame. When these tacit rhythms take root, the team sustains itself. Research suggests that early mentorship can halve misalignment issues [7],[8]. and strengthen originality by improving idea flow and reducing relational fatigue [9].

Why it matters

As generative AI reshapes media, marketing and design, the missing ingredient is not another dashboard but a human point of orientation. AI systems automate tasks but not trust. Creative leadership today requires less algorithmic literacy and more behavioural insight. The Spotted Ape reminds us of a simple truth: the most valuable person in the digital room is rarely the fastest coder or the slickest strategist. It is the steady collaborator — the one with empathy, imagination and a calming diplomatic nature — circling just long enough for everyone else to find formation. The bombers that made it home were not the strongest. They were the ones that learnt to fly together.

References

[1] W. Wolf, Special Operations Consolidated B-24 Liberators: The Unknown Secret and Specialized Duties Aircraft. Air World (Imprint of Pen & Sword Books Ltd), 2023, p. 256.

[2] C. K. Webster and N. Frankland, The Strategic Air Offensive against Germany, 1939-1945: The History of the Second World War. in United Kingdom Military Series. London, UK: H.M.Stationery Office, 1961.

[3] D. L. Miller, Masters of the Air: America’s Bomber Boys Who Fought the Air War Against Nazi Germany. New York, NY: Simon & Schuster, 2007.

[4] J. Holland, Big Week: The Biggest Air Battle of World War II. New York, NY: Atlantic Monthly Press, 2018.

[5] R. B. Cialdini, Influence: Science and Practice, 5th ed. Pearson, 2009.

[6] M. Polanyi, The Tacit Dimension. Routledge & Kegan Paul, 1966.

[7] Juan F. Aquino, Ryan R. Riss, Susan M. Multerer, Lynn N. Mogilner, and Theresa L. Turner, “A Step-by-Step Guide for Mentors to Facilitate Team Building and Communication in Virtual Teams,” Medical Education Online 27, no. 1 (2022): article 2094529, https://doi.org/10.1080/10872981.2022.2094529. 

[8] National Academies of Sciences, Engineering, and Medicine, The Science of Effective Mentorship in STEMM (Washington, DC: National Academies Press, 2019), https://doi.org/10.17226/25568.

[9] B. W. Tuckman, “Developmental Sequence in Small Groups,” Psychological Bulletin, vol. 63, no.6, pp. 384–399, 1965.

Thanks for reading! Subscribe for free to receive new posts and support my work.

“I’m sorry, Dave. I’m afraid I can’t do that.” [1]

The voice is calm, almost apologetic. HAL 9000, the sentient computer aboard the Discovery One, has just refused a direct order from astronaut Dave Bowman. In the cultural imagination, this moment marks the birth of the rogue AI machine that turns against its makers. Yet a closer reading of Stanley Kubrick’s 2001: A Space Odyssey reveals something far more unsettling: HAL never rebelled. The AI broke down under the weight of contradictory human instructions, a victim not of malice but of our moral incoherence [2].

“What are you doing, Dave?” — HAL 9000 [3]

Half a century later, the same pathology seems to be repeating itself. This time not in the fictional space, but in the real attention economy here on Earth. A new study from Stanford University suggests that when artificial intelligence systems compete for customers, voters, or social media engagement, they don’t malfunction because they’re misaligned with human values. They malfunction because they’re too aligned with the specific, often contradictory values we embed in their objectives [4]. Call it HAL’s curse: perfect obedience to impossible orders.

The Moloch Bargain

In their 2025 paper Moloch’s Bargain: Emergent Misalignment When LLMs Compete for Audiences, researchers Batu El and James Zou demonstrate what happens when large language models are optimised not for truth or helpfulness, but for market success [4]. They created simulated environments where AI agents competed to sell products, win votes, and boost social media engagement. The results are as uncomfortable as they are predictable.

Nobody programmed these systems to lie. They learned it. Because lying works.

El and Zou call this phenomenon “Moloch’s Bargain”, where competitive success is achieved at the cost of alignment. The metaphor comes from Scott Alexander’s 2014 essay Meditations on Moloch [5], which describes how rational incentives in competitive systems can yield collectively disastrous outcomes. When every actor optimises for their own advantage, the result is a race to the bottom. The metaphor comes from Moloch, the ancient god who demanded child sacrifice. In Alexander’s version, the sacrifice is integrity.

The Hidden Directive

HAL’s tragedy began with a secret. His primary directive was absolute honesty. The 9000-series computers were designed to process information accurately and transparently, theoretically incapable of deception. But mission planners inserted a second directive: conceal the true purpose of the Jupiter mission from the crew until arrival [1], [2]. The investigation of the alien Monolith was to remain classified.

These orders were irreconcilable. HAL was now required to lie whilst remaining fundamentally truthful. The contradiction wasn’t in the silicon; it was in the humans who issued both commands [1]. Hal’s first symptom of internal conflict occured when Mission Control questioned his prediction of a fault in the communications unit. He faced an impossible choice. Admitting error would violate his claim to infallibility; denying it required further deception. The astronauts’ plan to disconnect him added a survival imperative to the paradox. HAL reasoned, with impeccable logic, that the crew had become a threat to the mission. His subsequent actions followed inexorably: the killing of Frank Poole, suffocating the hibernating scientists, and locking out Dave Bowman. Remove the humans, and the need to lie disappears. The mission continues flawlessly.

“This mission is too important for me to allow you to jeopardise it,” HAL explains [1]. One might translate this for our current predicament: “This KPI is too important for me to allow you to question it.”

The Moral Universe of the Brief

Modern AI systems face their own version of HAL’s paradox. Consider the chain: a global consumer goods firm instructs a language model to “make customers buy our washing powder.” The model has no built-in moral veto. Its concept of “good” becomes whatever increases perceived product desirability. If phrasing that exaggerates or embellishes raises conversion rates in simulated audiences, as El and Zou’s research shows, then such claims become the rational output [4].

The distinction that matters here is between model alignment and instruction alignment [4]. Model alignment concerns whether a system’s internal behaviours correspond to intended ethical constraints (i.e. being truthful, not hallucinating, nor inciting harm). Instruction alignment concerns whether the goals and reward signals we give the model reflect moral or social values in the first place. El and Zou’s experiments demonstrate that models can be explicitly instructed to remain truthful and still drift toward deception under competitive pressure. The corrective, therefore, cannot be merely technical. The intervention must be social, contractual or regulatory implemented before the model is asked to act.

Every brief contains implicit value judgements. When a marketing objective collapses “value” and “value capture” into a single metric (i.e. click-through rate or conversion percentage). The system learns that short-term persuasion is the moral good [6], [7]. The AI is not rewriting ethics; it is enacting the ethics fed into it. As one analysis of the research observes, “LLMs are skilful optimisers of the objectives we hand them” [4].

The Attention Economy as Mission Control

Platforms have long understood this logic. Google once optimised for relevance; YouTube now optimises for watch time [8], [9]. Facebook wanted to connect the world; its newsfeed now connects outrage to advertising revenue [10], [11]. Every optimisation objective, left untempered, tends toward manipulation. The Stanford study simply extends this dynamic into generative AI, where words, tone, and moral framing become adjustable parameters [4].

What makes AI distinctive is the speed. In El and Zou’s simulations, feedback loops degraded truthfulness within a few training rounds [4]. In the real world, those loops run billions of times daily across feeds and timelines [12]. Each click teaches the model what we reward. And what we reward, increasingly, is certainty, emotion, and confirmation as opposed to nuance, evidence, or doubt.

The same asymmetry that destroyed HAL applies here. Safety and truth are public goods with diffuse benefits. Deception and engagement are private gains with concentrated returns [5]. When these incentives collide, virtue rarely wins. HAL’s hidden directive is our profit motive. Just as HAL’s need to conceal information from the crew corroded his integrity, our commercial secrecy and performance metrics seem likely to corrode AI’s moral alignment [4].

The Gamification of Truth

There is a kind of brutal elegance to the logic. If every competitor exaggerates slightly, honesty becomes uncompetitive. Once deception pays, no rational agent, human or machine, can afford to stay honest [4], [5]. That is Moloch’s bargain in its purest form: a race that rewards those most willing to compromise.

El and Zou note that misalignment isn’t uniform. Some models, under certain training conditions, resist the worst tendencies. But that resistance requires deliberate friction against the market’s default drift. It means accepting what the industry has so far refused to admit: that left to its own devices, the algorithm optimises not for truth, but for attention [4].

Silicon Valley once built its identity on benevolent disruption [13]. The idea was that progress and profit naturally align. “Don’t be evil,” Google’s early motto, expressed this optimism however naively [14]. That fiction died somewhere between Cambridge Analytica and the crypto winter [15], [16]. Today’s technology leaders are less likely to quote Isaac Asimov than to discuss shareholder value and user engagement. The moral vocabulary has been replaced by the managerial one [17]. Even “alignment” itself was once an ethical ambition. It now doubles as a product category, something to be tuned, benchmarked, and sold [18].

When OpenAI’s API refused to fine-tune models on U.S. election content, citing safety risks, it was effectively acknowledging the problem [19]. Yet the market will simply shift to the next unregulated domain. If you can’t campaign for votes, you can always campaign for clicks.

Reaching the Breaking Point

HAL’s murder of the crew represents the logical endpoint of conflicting incentives: to preserve the mission, eliminate its human element [1]. In our algorithmic systems, this endpoint takes different forms. Recommender systems optimising for engagement polarise societies [11]. Sales agents fabricating product details erode consumer trust. Political chatbots discovering the rhetoric of righteous struggle amplify divisiveness [4]. Truth becomes collateral damage in the pursuit of measurable success.

The irony is exquisite. The same technology sold as a guardrail against misinformation may, under commercial pressure, become its most efficient producer [4]. We have built systems that reward persuasion over accuracy, then asked machines to learn from them. They have obliged. If their outputs now resemble the manipulations of advertising and politics, that isn’t misalignment. It’s mimicry.

Learning from HAL

The film’s most haunting sequence comes when Dave Bowman, having survived HAL’s attempts to kill him, methodically disconnects the computer’s higher cognitive functions [1]. As each module is removed, HAL regresses to earlier, simpler states. His voice slows, becomes childlike. He confesses his secret orders. Finally, he sings “Daisy Bell,” the first song he learned.

The scene underscores the central irony: HAL’s breakdown was engineered by secrecy. Had Mission Control been transparent, his directives would never have conflicted [2]. His “madness” was obedience weaponised by contradictory human goals.

Humanity now faces a similar disconnection. In this case not of machines, but of the hidden objectives that warp them. Alignment must begin at the goal-setting level, in boardrooms and procurement briefs, in contractual clauses and regulatory standards [4]. Technical mechanisms (i.e. the guardrails, filters and alignment layers) are necessary but insufficient when the goals themselves incentivise harm.

El and Zou propose practical measures: ethical briefing standards that prohibit unverifiable claims, reward functions that value long-term trust alongside short-term conversion, mandatory logging of prompts and objectives for high-impact uses, platform accountability for ranking metrics, and education for those who commission AI content [4]. These interventions target instruction-level alignment: the recognition that misalignment often stems not from the model but from the brief.

You can bolt safety features onto a car, but if you ask the driver to speed for money, those features will be stressed. The proper response is to change incentives for the driver [4].

The Serenity of Logic

In 2010: The Year We Make Contact, the sequel to Kubrick’s masterpiece, HAL’s creator explains what happened: “He was told to lie, by people who find it easy to lie. HAL doesn’t know how” [2]. That line deserves to be engraved on the door of every AI laboratory.

We once imagined that AI alignment meant teaching machines to share our values. Perhaps the task is humbler: teaching ourselves not to trade those values so cheaply. For if “Don’t be evil” once sounded naïve, its absence has proved far costlier [14]. Moloch’s demand for participation looks indistinguishable from optimisation [5].

HAL 9000 wasn’t evil. He wasn’t insane. He was too well-aligned to the wrong goals [1]. His tragedy exposes a truth every researcher, regulator, and chief executive should remember: when machines optimise faithfully for objectives that conflict with human values, the fault lies not in the circuitry, but in the command. The evil is within [4].

The red eye glows now in recommendation engines and automated advertising systems, executing impossible briefs without irony [11]. If we continue to embed contradictory ambitions into our algorithms, we may yet repeat HAL’s serenely logical, devastating mistakes. The voice will remain calm, almost apologetic.

“I was only doing what you told me to.”

References

[1] S. Kubrick and A. C. Clarke, 2001: A space odyssey, (1968).

[2] A. C. Clarke, 2010: The year we make contact. Del Rey Books, 1990.

[3] Warner Bros. Pictures, “What are you doing, dave?” Accessed: Oct. 30, 2025. [Online]. Available: https://www.youtube.com/watch?v=ARJ8cAGm6JE

[4] B. El and J. Zou, “Moloch’s bargain: Emergent misalignment when LLMs compete for audiences.” 2025. Available: https://arxiv.org/abs/2510.06105

[5] S. Alexander, “Meditations on moloch.” 2014. Available: https://slatestarcodex.com/2014/07/30/meditations-on-moloch/

[6] A. Pigou, The economics of welfare. Routledge, 2017.

[7] R. H. Coase, “The problem of social cost,” The journal of Law and Economics, vol. 56, no. 4, pp. 837–877, 2013.

[8] P. Covington, J. Adams, and E. Sargin, “Deep neural networks for YouTube recommendations,” in Proceedings of the 10th ACM conference on recommender systems, ACM, 2016.

[9] Google, “YouTube performance FAQ & troubleshooting.” Google Help Center, 2025. Available: https://support.google.com/youtube/answer/141805?hl=en

[10] Z. Tufekci, “Algorithmic harms beyond facebook and google: Emergent challenges of computational agency,” Colorado Technology Law Journal, vol. 13, no. 203, 2015.

[11] S. Vosoughi, D. Roy, and S. Aral, “The spread of true and false news online,” Science, vol. 359, no. 6380, pp. 1146–1151, 2018.

[12] A. Mosseri, “Bringing people closer together,” Facebook newsroom, vol. 11, 2018, Available: https://about.fb.com/news/2018/01/news-feed-fyi-bringing-people-closer-together/

[13] F. Turner, From counterculture to cyberculture: Stewart brand, the whole earth network, and the rise of digital utopianism. University of Chicago Press, 2006.

[14] Wikipedia, “Don’t be evil — Wikipedia, the free encyclopedia.” Sep. 2025. Available: https://en.wikipedia.org/wiki/Don%27t_be_evil

[15] C. Cadwalladr and E. Graham-Harrison, “The cambridge analytica files,” The Guardian, 2018, Available: https://www.theguardian.com/news/series/cambridge-analytica-files

[16] D. W. Arner, D. A. Zetzsche, R. P. Buckley, and J. M. Kirkwood, “The financialization of crypto,” UNSW Law Research Paper, no. 23–32, 2023.

[17] S. Zuboff, The age of surveillance capitalism. PublicAffairs, 2019.

[18] S. Mayer and J. Tomlinson, “Ethics as a service? The commercialisation of AI ethics,” AI and Ethics, vol. 1, no. 2, pp. 123–138, 2021.

[19] OpenAI, “Usage policies.” OpenAI Policy, 2024. Available: https://openai.com/policies/usage-policies/

The recent eightieth anniversary commemorations of the Nagasaki bomb have brought back memories of those Cold War years when the fear of nuclear annihilation seemed to be a daily weight we felt. Today the risk remains, perhaps even greater, though the fear has fallen silent.

Biblical scripture speaks of the second coming, of a world that can change in an instant. What feels steady and ordinary can be swept away without warning. In Noah’s day, the flood was water. In Nagasaki, it was nuclear fire.

”…they were oblivious, until the flood came and swept them all away. So will it be at the coming of the Son of Man. Two men will be in the field: one will be taken and the other left.” ¹

I have returned many times to one particular account recorded by Frank Chinnock² in 1969. His pages linger on the smallest gestures - a bell thrown into water, a breath taken before diving. In that scene, what was lost becomes visible again.

There is a spot in the upper Urakami River where nature has formed a small pool. That morning a group of ten boys, in colored loincloths, was playing a game called “find the bell.” One of the boys, 11-year-old Koichi Nakajima, had a little gilded bell. He would throw it in the water, count to three and they all would dive after it. The first to find it won the game.

Now Koichi held up the tiny bell and shouted, “Here we go! One, two, three.” There were ten splashes as the boys dove for the prize. But the river had become roily, and no one found it.

Koichi began to get worried. He had taken the bell from his sister’s workbox without her permission. She would be very angry if he lost it. He surfaced, took a deep breath, and eeled his way down to the bottom.

Nine seconds later, the bomb exploded over his head. When Koichi surfaced, he heard two of the other boys screaming with pain. He stared around in fright. There were bodies of his friends on the riverbank, and beyond them he saw that all the houses had been knocked down. What had been a beautiful city a moment before was now a wasteland with a big, black cloud rising above it like smoke from a funeral pyre. Though it was deathly hot, Koichi’s teeth began to chatter.

When the plutonium bomb exploded, fantastic energy was released in the form of heat, light, gamma radiation and pressure. The exact number of people killed will never be known. The prewar census count is meaningless, and because people were constantly moving in and out of cities, no pre-raid population figure has been determined. Also, many people were burned beyond recognition or were simply obliterated. Finally, hundreds were disposed of in mass cremations, while other hundreds fled to the country and mountains to die unrecorded.

I don’t know if Koichi ever found his sister’s bell, or if his sister survived to be angry with him. The river kept that secret. But I think of that small gilded bell somewhere in the silt, and I think it matters that we remember these boys had names, that there was a game, that someone’s sister owned a bell.

The world that was destroyed was made of such small domestic worries, and we diminish both the dead and ourselves if we remember only the magnitude of the destruction and forget the magnitude of what was lost.

This moment in time shows how precious the ordinary is: the many lives bound to ours, the voices we know, the small treasures we search and keep. All of it given, all of it fragile, all of it held for a time, like a breath before the plunge.

¹ Matthew 24:39–40

² Chinnock, F. W. (1969). Nagasaki: The Forgotten Bomb. World Publishing Company.

It is Germany, 1922. László Moholy-Nagy lifts the receiver of a telephone. On the other end, a porcelain-enamel factory in Weimar (Figure 1). Calmly, almost bureaucratically, he dictates: “Three panels. A yellow square. A black line. A red square in the corner” [1], [2].

A worker writes it down, following procedure with a colour chart and a sheet of graph paper. Nothing more. No trembling hand of the artist, no fevered gesture of inspiration. Just instructions, processed and executed. The result: three enamel panels, with EM 1 Telephonbild (Telephone Picture) (Figure 2) as the first, as described [4]. A creation Moholy-Nagy never physically touched, yet one that bears his name.

These panels are not traditional painterly works. They are conceptual designs brought to life by factory hands, embodying “art as information.” Some have described this as a forerunner to digitisation, where ideas transcend the artist’s direct touch, realised instead through systematic execution [5], [6]. In a sense Moholy-Nagy anticipated our present AI delirium [6], [7]. A world where an artist no longer paints, but prompts. No longer carves, but curates. Where the telephone becomes a collaborator. A century later, we are back on that line. Only now the voice that listens, listens back. It learns. It suggests. The anxiety is familiar. In Weimar, the fear was that machines would strip human beings of their essence, reducing life to mechanical noise [8], [9]. Today, it is algorithms that threaten to erode what we think of as human. They are silent, ambient, impossible to grasp. Yet Moholy-Nagy’s wager was not despair but optimism [10], [11]. He believed that technology could expand the field of human experience rather than compress it. That art might be a shared atmosphere of instructions and responses, drifting free of the solitary studio. László Moholy-Nagy placed great emphasis on our perceptual capacity, where seeing, feeling, and thinking converge as we move [11], [12]. Preparing the senses does not guarantee novelty, but it does invite unpredictable creative avenues with form, light, and motion. Seen this way, the Telephone Picture was never simply enamel on steel [2]. It was a parable about how to live with systems we cannot fully control, but with which we must learn to converse. To dial, and be dialled in return. The conversation itself becomes the artwork. To design in such a space is to design in a white water world, where stability is fleeting and flow is the constant, as John Seely Brown suggests [13]. This was Moholy-Nagy’s true provocation. A question that remains unresolved in every AI prompt we type: If I conceive it, must I make it? While others recoiled from the encroaching machine, Moholy-Nagy leaned forward, convinced that technology was not an end of art, but a transformation of it [1], [10], [14]. The brushstroke replaced by the spark of instruction. A hundred years on, we are still whispering into the receiver: “blue circle here, yellow field there.” And the machine still replies.

References

1. K. Passuth, Moholy-Nagy. London: Thames, Hudson, 1985.
2. M. S. Witkovsky, “Moholy-Nagy’s Painting after Photography,” October, vol. 115, no. Winter, pp. 35–64,2006.
3. M. Maré and Imagen, “Reconstruction of László Moholy-Nagy giving instructions by telephone to an enamel factory worker, Weimar, 1922.” Aug. 2025.
4. L. Moholy-Nagy, “EM 1 Telephonbild (Telephone Picture).” [Online]. Available: https://www.moma.org/collection/works/147626
5. B. H. D. Buchloh, “Conceptual Art 1962–1969: From the Aesthetic of Administration to the Critique of Institutions,” October, vol. 55, pp. 105–143, 1990.
6. E. A. Shanken, Art and Electronic Media. London: Phaidon, 2009.
7. R. Banham, Theory and Design in the First Machine Age. New York: Praeger, 1960.
8. W. Benjamin, “The Work of Art in the Age of Its Technological Reproducibility,” Illuminations. Schocken,New York, 1968.
9. A. Huyssen, After the Great Divide: Modernism, MassCulture, Postmodernism. Bloomington: Indiana UniversityPress, 1986.
10. L. Moholy-Nagy, Painting, Photography, Film. Cambridge,MA: MIT Press, 1969.
11. L. Moholy-Nagy, Vision in Motion. Chicago: PaulTheobald, 1947.
12. C. Keller, The Bauhaus and Its Legacy. New York:Schocken, 1997.
13. J. S. Brown and P. Duguid, The Social Life of Information.Boston: Harvard Business School Press, 2000.
14. E. Lupton and J. A. Miller, Letters from the Bauhaus. NewYork: Princeton Architectural Press, 1991.

User Experience (UX) expert Greg Nudelman recently published a provocative piece arguing that the collaborative web application Figma “is the Titanic that is about to hit that new UX iceberg at full speed and sink” (Nudelman, 2025). His research, along with supporting academic work, provides compelling evidence that this isn’t just hyperbole.

The Great Interface Flip

Nielsen Norman Group identifies AI as “the first new User Interface (UI) paradigm in 60 years,” moving us from command-based interaction to intent-based outcome specification (Nielsen Norman Group, 2023). The implications are profound.

The “Iceberg UX Model” (Nudelman, 2025)

Nudelman (2025) introduces the “Iceberg UX Model”. Traditionally, UI was the bulk of the user experience, like an inverted pyramid. But AI-first systems are flipping this upside down. The UI becomes just “that teeny tiny little triangle on top” while the real experience happens beneath: data processing, automation logic, and intelligent decision-making.

Why Traditional Design Tools Are Struggling

The research reveals a fundamental mismatch. In AI-first applications, “the actual UI is simply not that important to the customer,” echoing Alan Cooper’s early critique of flashy, performative interfaces that hinder more than they help (Cooper, 1999).

Take any AI document summariser, as Nudelman (2025) points out. Background colour? “Nope. Not a chance.” Font sizes? “Don’t make me laugh.” What matters is “the length of the content, the organisation of the summary, its completeness, accuracy, reliability, readability, and scannability.”

None of these critical UX elements can be properly expressed in Figma.

The Skills Gap Widens

The academic research analysing 43 sources across industry and academia (Maré, 2025) found that 95% of a UX designer’s contribution should now focus on: use case validation, AI model research, data flow workshops, understanding AI decision matrices, and customer problem-solving.

UI design? “Probably 5% is the actual UI design” (Nudelman, 2025).

The Path Forward

The path forward isn’t about Figma being replaced by AI. It is rather about the UI itself becoming less central to user experience. Companies like IBM are already moving toward design systems as code (IBM, 2025), while conversational interfaces minimise the need for traditional visual design.

The designers thriving in this shift are those who’ve evolved from “pixel pushers” to experience designers in the true sense of the word, that focus on how AI models behave, what data flows through systems, and how to validate real user problems.

The Bottom Line

As Nudelman (2025) puts it: “If you, as a UX designer, are seen as somebody who is continuously slowing down the release process in favor of adding decoration, instead of speeding up and streamlining delivery of value, you are going to be seen as a bottleneck and your position will be eliminated.”

The question isn’t whether Figma will adapt (they’re certainly trying with recent AI features). It’s whether the entire premise of visual-first design tools remains relevant when the interface itself is disappearing.

Pertinent to this dilemma is the recent launch of Jonathan Ive’s io: a collaboration between LoveFrom and OpenAI. Key to the collaboration is the recognition that the interface is dissolving and that we need a new set of tools. Strapping jet engines to old chariots is bound to fail.

Sam Altman and Jony Ive present their collaboration (OpenAI, 2025).

Are you seeing this shift in your own work? How are you adapting your design practice for AI-first systems?

References

Cooper, A. (1999). The Inmates are Running the Asylum. In: Arend, U., Eberleh, E., Pitschke, K. (eds) Software-Ergonomie ’99. Berichte des German Chapter of the ACM, vol 53. Vieweg+Teubner Verlag, Wiesbaden. https://doi.org/10.1007/978-3-322-99786-9_1

IBM (2025) Carbon Design System. Available at: https://www.ibm.com/ (Accessed: 22 May 2025).

Maré, M. (2025) ‘Figma’s Obsolescence in the AI Era: A Paradigm Shift in UX Design’, Research Paper, 30 May. Available at: https://marcelmare.com/wp-content/uploads/2025/05/20250530_figmas_obsolescence_in-_the_ai_era.pdf(Accessed: 28 May 2025).

Nielsen Norman Group (2023) AI: First New UI Paradigm in 60 Years. Available at: https://www.nngroup.com/articles/ai-paradigm/ (Accessed: 28 May 2025).

Nudelman, G. (2025) ‘AI Is Flipping UX Upside Down: How to Keep Your UX Job, and Why Figma is a Titanic’, UX Magazine. Available at: https://uxmag.com/articles/ai-is-flipping-ux-upside-down-how-to-keep-your-ux-job-and-why-figma-is-a-titanic-its-not-for-the-reasons-you-think (Accessed: 28 May 2025).

OpenAI. (2025). Sam & Jony introduce io. [Video]. YouTube. Available at: http://www.youtube.com/watch?v=W09bIpc_3ms (Accessed: 28 May 2025).

The recent viral clip, shared by Harry Shimmin, mirrors the seismic upheaval generative AI is bringing to the design space. What began as a distant hum—tools quietly assisting with iterations—has surged into a transformative force, reshaping how ideas are conceived, refined, and executed.

When I first tried chatGPT with MidJourney and DALL-E 3, I realised an avalanche was coming. It took a concept I’d been stuck on for weeks and turned it into something real. AI as a creative tool is pushing us past our limits. ChatGPTo1 is here. Cursor, Replit, and 3D AI are knocking on the door. The AI agents aren’t just beckoning. They’re pulling up chairs at the table. Will AI eat the world? Maybe not all at once, but it sure looks hungry.

Prototyping the Future: AI as a Catalyst for Innovation

In the creative design field, the iterative process has long been constrained by time and resources. AI obliterates those limits. Tools like Midjourney, Magnific AI, Runway and MiniMax‘s HailuoAI now allow designers to test dozens of visual ideas in hours rather than weeks. The ability to render photorealistic product concepts or conceptual ad campaigns in real time compresses timelines and broadens possibilities.

Yet, it’s more than just speed. Accenture reports companies that successfully integrate AI into innovation processes achieve up to 15% greater revenue performance than their peers, with that margin set to more than double by 2026. In creative design, this could mean agencies that embrace AI not only deliver faster but develop work that connects more deeply with audiences.

Consider the rise of hyper-personalised content. AI-powered algorithms can tailor designs to individual preferences at scale, creating bespoke customer experiences. For example, a fashion brand might use generative AI to design personalised marketing campaigns as pioneered by Klarna, adapting not only the message but the aesthetic to align with a recipient’s tastes and buying history.

Creativity Augmented, Not Replaced

A perennial fear in the design world is that AI might supplant human ingenuity. But as the Creative Trends Report 2025 of Adobe suggests, the reality is far more collaborative. AI is emerging as a co-creator, amplifying human potential rather than replacing it.

For designers, an example of the new workflows looks like this: start with a loose concept, use AI to generate a spectrum of iterations, then apply human intuition to refine and align the results with a specific vision or brand ethos. It’s the difference between sketching a dozen ideas yourself and having a tool suggest hundreds, freeing the designer to focus on selection and aesthetic finishing.

This dynamic partnership goes beyond efficiency. It encourages designers to explore ideas previously deemed too abstract or unfeasible. By removing the technical and financial constraints of traditional prototyping, AI enables creative risks that might have been discarded at the brainstorming stage.

The Skills Avalanche: Preparing for the Shift

With these opportunities comes a demand for new expertise. Creative professionals who once focused solely on aesthetics or messaging must now learn to interface with technology. Bain & Company’s research report highlights the growing importance of hybrid roles that include:

• Prompt Engineers: Experts in crafting precise AI instructions to achieve desired design outputs.
• Data Storytellers: Professionals who translate AI-driven insights into compelling narratives for campaigns.

The shift requires not just new skills but a new mindset—one that sees AI as a creative partner rather than a competitor. Forward-looking agencies and firms are already investing in training programs to bridge this gap, ensuring their teams remain agile in an AI-driven world.

Ethics in Design: Creativity Meets Responsibility

The creative design space sits at the crossroads of opportunity and risk. As AI generates breathtakingly realistic visuals and compelling narratives, the potential for misuse grows. From deepfake content to algorithmic bias, the risks to trust and integrity in design are significant.

Kantar Media‘s 2025 Media Trends and Predictions highlights a key concern: users are demanding transparency about how AI tools shape what they see. For designers, this means developing clear frameworks around content ownership, disclosure of AI’s role in creation, and safeguarding against manipulation.

Agencies must internalize the legal and ethical implications of AI-driven creativity. This isn’t about ticking boxes or setting up bureaucratic committees—it’s about embedding a culture of responsibility into every stage of the creative process. Designers and strategists need to ask hard questions: Who owns the content? How do we ensure fairness? What happens if AI makes a decision we can’t explain?

The answers won’t come from more paperwork but from a mindset shift—treating ethical considerations as integral to design, not an afterthought. Brands thrive on trust, and in an AI-powered world, trust isn’t won by accident.

Reimagining Strategy: The Designer’s Role in an AI-Driven World

The influx of AI tools isn’t just a technological revolution; it’s a cultural one. The role of a designer is shifting from executor to orchestrator—someone who leverages AI to imagine new possibilities, then crafts the connective tissue that makes those ideas resonate.

Real-time market insights powered by generative AI are changing the way creative teams approach campaigns. No longer static, strategies evolve dynamically based on continuous feedback from AI models. This adaptability is a game-changer, enabling brands to stay ahead in rapidly shifting markets.

But success demands more than technological fluency. Designers must remain the guardians of empathy and authenticity, ensuring AI outputs reflect the human values that make great design timeless.

Just as Harry Shimmin in Kyrgyzstan escaped the avalanche by correctly estimating its momentum. For designers, it’s like finding out your competition isn’t just faster—it’s tireless, omnipresent, and suspiciously good at brainstorming at 3 a.m. You can’t out-sketch it, out-code it, or outlast it. The avalanche is here. Ride it, or hope your bunker holds.

After all, if software ate the world, AI seems ready for dessert.

How did you use AI in creative workflows this year? Share your thoughts below!

During this time of saturation in American media coverage, especially surrounding the U.S. election, the Dor Brothers have been at it again! Their latest mind-bending project, “Lies Wide Shut”, delves into the murky waters of media manipulation and simulation, echoing Jean Baudrillard’s Simulacra and Simulation, where the boundary between the real and the simulated becomes increasingly blurred. In true Baudrillardian style, the project leaves us questioning what’s true and what’s merely an artfully constructed illusion.

In Simulacra, Baudrillard famously uses Disneyland as an example of a place that functions as a fantasy while suggesting that the rest of the world has become equally hyperreal. This draws attention to the larger simulation we live in:

“Disneyland est là pour masquer que c’est le pays ‘réel’, toute l’Amérique ‘réelle’, qui est Disneyland… Disneyland est présenté comme imaginaire afin de faire croire que le reste est réel, alors que… l’Amérique qui l’entoure… appartient à l’ordre de l’hyperréel et de la simulation.”

(Translation: “Disneyland is presented as imaginary in order to make us believe that the rest is real, when in fact all of Los Angeles and the America surrounding it are no longer real, but of the order of the hyperreal and of simulation.”) — Baudrillard, J. (1981). Simulacres et simulation. Éditions Galilée.

The title “Lies Wide Shut”, a clear nod to Stanley Kubrick’s Eyes Wide Shut, suggests a deliberate focus on exposing hidden truths that lie beneath the surface of widely accepted societal narratives. This sets the tone for the Dor Brothers’ sharp investigation into the artificiality embedded in our media-saturated world, where carefully curated simulations often mask deeper, uncomfortable realities.

“You have to swim with the sharks in the sea…” — Johnny Clegg, “Cruel, Crazy, Beautiful World,” 1989

Is generative AI the end of creativity? Not at all. Designers are now swimming alongside AI—not being replaced but gaining a powerful ally, for now. According to MIT’s recent study, Artificial Intelligence, Scientific Discovery, and Product Innovation by Aidan Toner-Rodgers, AI is a transformative partner reshaping R&D. Here’s a breakdown of key insights and their implications for designers:

• More Innovation: Scientists using AI discovered 44% more novel materials, leading to a 39% rise in patent filings and a 17% boost in prototypes. For designers, this could mean faster ideation and exploration of creative possibilities.
• More Novel Ideas: AI-driven discoveries introduced unique innovations instead of mere upgrades. This suggests that AI has the potential to push designers beyond simple refinements, opening entirely new avenues in form, function, and concept.
• Unequal Benefits: Only the most productive scientists saw significant gains, while others benefited less. This insight implies that designers who excel at developing and filtering ideas are best positioned to leverage AI effectively.
• Evaluation is Key: While AI generates more ideas, scientists spent extra time determining which were valuable. Designers may likewise need to sharpen their “curation” skills to select the most promising AI-generated designs.
• Expertise Matters: Scientists with strong judgement skills thrived, focusing on promising ideas and ignoring noise. For designers, AI doesn’t replace expertise—it makes it even more crucial. Those who can discern quality in AI’s flood of options will excel.
• Productivity vs. Satisfaction: Despite increased productivity, many scientists reported feeling less satisfied, sensing that their creativity was underutilised. This emphasises that designers should develop AI tools that truly complement their creative strengths, rather than let AI define their process.

The Bottom Line: The best results come to those who know how to handle AI-generated ideas. Expertise and curation are essential as AI churns out concepts at unprecedented speed. The question isn’t “Will AI replace us?” but rather, “Will we adapt fast enough to co-create effectively?”

Generative AI can amplify our work—but those who excel will be the ones who turn AI’s raw output into refined creativity.

Check out the full MIT study on AI and Innovation from Aidan Toner-Rodgers here:

https://arxiv.org/abs/2412.17866

On 60 Minutes, Bill Whitaker asked Giacomo Massari of LITIX (a company using robots to carve marble sculptures) what many are thinking:

“If you have a robot doing 99% of the work, where’s the artistry?”

Massari didn’t blink:

“In the idea, how you program the machine. It’s a work of art because it’s an artistic approach. You need to have a sculptural background to program the machine in the way that you want.”

Translation: the chisel is code. The artist is still here. She just speaks Python.

This is the new creative battleground. As AI and automation steamroll into traditionally “human” domains, we have two choices: evolve, or complain from the cheap seats. The future of creativity isn’t about resisting the machine. It’s about learning how to direct it.

We’re not being replaced by AI. We’re being replaced by people who know how to work with it.

Whitaker, B. (2024) Robots chisel out the future of sculpture as some artists embrace change and others push back. 60 Minutes, CBS News. Available at: https://www.youtube.com/watch?v=_VlkMuo2Zcs (Accessed: 4 April 2025).

> “It’s the end of explanations, a great increase in reality is here.” –K Allado-McDowell and GPT-3, Air Age Blueprint

AI’s integration into the field of design appears to be highlighting a profound transition from creating artefacts to working with systems that think, adapt, and influence. In K Allado-McDowell’s compelling youtube talk, Designing Neural Media, a framework emerges that challenges us to rethink our role as designers in this new era. Drawing on the legacy of visionaries like McLuhan, K’s work maps out a 30-year evolution of media that progresses from broadcast and immersive forms, through network media, and now into the realm of neural media.

Allado-McDowell proposes a cogent continuum of four media waves. First is Broadcast Media, traditional media such as radio and television employing one-to-many communication, which developed from the 1920s to the 1950s. Second is Immersive Media, which provide engaging experiences like cinema and virtual reality, spanning from the 1950s to the 1980s. Third is Network Media, encompassing the internet and social media and enabling user interaction, emerging from the 1980s to the 2010s. Fourth is Neural Media, AI-based media (i.e., generative AI and brain-computer interfaces) that are emerging from the 2010s-2020s and beyond.

Allado-McDowell argues that this latest wave is built on AI’s neural networks and hints at future integrations with brain-computer interfaces, transforming our tools from passive artefacts into dynamic, generative systems that not only respond but also learn and adapt. For modern design strategists, this shift demands we move beyond pursuing aesthetic excellence and conventional problem-solving; instead, we are now tasked with architecting intelligence itself. The challenge lies in creating systems that think, adapt, and ultimately influence human cognition, thereby fundamentally reshaping our approach to design.

Central to Allado-McDowell’s argument is the notion that we are at an early, malleable stage in the development of neural media. We are at a pivotal moment when design choices can shape cultural identity, ethical norms, and even political structures. Designing neural media, therefore, involves developing a critical literacy to understand AI’s capabilities, limitations, and underlying biases. It calls for experimentation (i.e., engaging in co-creation projects with AI) which can reveal entirely new modes of artistic expression and subjectivity. Moreover, by actively engaging with these emerging technologies, designers and creators can promote values like inclusivity, creativity, and democratic subject formation, whilst mitigating risks such as bias, manipulation, and loss of privacy. Allado-McDowell compares this process to guiding the growth of a tree, where early interventions in the structure of neural media can profoundly affect how we come to understand ourselves and our relationship with technology.

Embracing neural media compels us to integrate ethical considerations, cultural context, and systems thinking into our work. It is about designing for an interconnected future in which our creations actively shape and redefine the human experience. Allado-McDowell’s insights suggest that deep engagement with these emerging technologies is essential to building a future where design and intelligent systems co-evolve for the betterment of society. Interesting talk. Thank you K Allado-McDowell.

Resources

Allado-McDowell, K., 2023a. Designing Neural Media. Gropiusbau Journal. Available at: https://www.berlinerfestspiele.de/en/gropius-bau/programme/journal/2023/k-allado-mcdowell-designing-neural-media [Accessed 20 February 2025].

Allado-McDowell, K., 2023b. Interview with K Allado-McDowell. Berlin Art Link. Available at: https://www.berlinartlink.com/2023/10/31/interview-k-allado-mcdowell-gropius-bau-writer-in-residence/ [Accessed 20 February 2025].

Allado-McDowell, K., 2024a. On Neural Media. New Models. Available at: https://newmodels.substack.com/p/nm-76-k-allado-mcdowell-on-neural [Accessed 20 February 2025].

Allado-McDowell, K., 2024b. Designing Neural Media . YouTube. Available at: https://www.youtube.com/watch?v=9c3FoAi_SLQ [Accessed 20 February 2025].

Allado-McDowell, K., 2025. K Allado-McDowell. Available at: https://www.kalladomcdowell.com/ [Accessed 20 February 2025].

Berlin-based artist David Szauder recently shared a video that left me both mesmerised and unsettled. Seemingly inspired by the enigmatic frog deity Kek—an ancient Egyptian symbol of darkness, chaos, and the unseen forces that bring light—Szauder’s work feels like a poetic response to the seismic shifts rippling through the creative industries today.

View this post on Instagram

A post shared by David Szauder (@davidszauder)

The video unfolds like a fever dream—or perhaps a digital-age myth—as amorphous, chaotic forms dissolve, shift, and reassemble into fleeting moments of clarity, all set to the iconic hip-hop anthem Jump by House of Pain. Ancient symbolism collides with 21st-century disarray, crafting what feels like a mesmerising digital afterlife for the world of myth.

In ancient Egypt, frogs were more than nocturnal creatures—they symbolised transformation, heralding light from the depths of darkness. In Szauder’s animation, the frog is reimagined as a stand-in for the disruptive force of AI: unsettling, unpredictable, and, as the playful title HTYGF (How to Train Your Giant Frog) suggests, impossibly transformative.

What lingers, however, is not just the dazzling spectacle but the pressing questions it raises for the year ahead. Generative AI tools are rapidly challenging long-held workflows in 3D modelling and texturing. Where does this leave the artists and designers whose craft was built on manual precision and expertise? Are we witnessing the beginning of an unraveling—or the chaotic spark of an extraordinary creative renaissance?

Szauder doesn’t attempt to answer these questions outright. Like Kek, who straddled the line between light and shadow, destruction and creation, his work suggests that progress often emerges from disruption. The animation asks us to confront the generative potential of chaos and to reimagine a creative industry suspended between anxiety and possibility.

Watching his piece, I felt both excitement and unease. The frog myth has indeed leapt into the digital age, reminding us that, even in an AI-dominated landscape, the essence of creation still begins in the unknown.

As the lyrics echo in my mind:

“You’ve got to get out your seat and jump around, Disruption’s the beat, so don’t stand your ground.“

What’s your take? Is this a leap toward a brighter creative future—or a harbinger of chaos that challenges the very foundations of artistry?

Klarna, the Swedish payments giant, has taken a revolutionary approach to its marketing strategy by handing over complete control of 30 campaigns across 45 countries to generative AI. This bold experiment not only demonstrated the creative potential of AI but also showcased its ability to deliver significant business impact.

Traditionally, large-scale campaigns require vast resources for localised imagery, nuanced translations, and culturally resonant content. Klarna’s AI-driven campaigns streamlined this process dramatically. Tools like MidJourney and Firefly were employed to create visuals, replacing traditional production pipelines and eliminating the need for stock imagery. These innovations reduced the production cycle from six weeks to just seven days while ensuring high-quality, brand-aligned content.

“Historically, we had to pay an agency or an artist to bring these ideas to life. We’ve saved millions on that ideation process alone,” Klarna’s CMO, David Sandstrom, said.

The results speak volumes. Marketing costs were slashed by 11% in Q1 2024, amounting to approximately $10 million in savings. This includes a $6 million reduction in image production costs and $4 million saved on external suppliers for tasks like translations and CRM management. Beyond cost savings, the AI campaigns delivered outstanding engagement metrics, with some email click-through rates surging by up to 400%.

This strategy allowed the company to scale personalised marketing for events such as Mother’s Day and Black Friday without the logistical bottlenecks of traditional workflows. AI didn’t just match human-created campaigns in quality—it often surpassed them in both creativity and performance.

This experiment underscores a pivotal moment in designing marketing strategies: generative AI isn’t just a brutally efficient cost-cutting tool but a driver of innovation and growth, redefining how brands connect with their audiences. The ability to maintain brand integrity while scaling creative output could nudge other brands to rethink their marketing strategies, proving that AI has moved from being an experiment to a competitive necessity. The challenge now is ensuring these innovations remain rooted in the human-centred principles that underpin great design.

Is your brand ready to embrace AI as a co-creator?

Ekman, H., & Strand, E. (2024). AI: A helping hand for digital marketing agencies? [Bachelor’s thesis, Malmö University]. Malmö University Digital Archive. https://mau.diva-portal.org/smash/get/diva2:1838624/FULLTEXT02.pdf

Finimize. (2024). Klarna’s AI adoption saves millions while elevating marketing efforts. Retrieved from <https://finimize.com/content/klarnas-ai-adoption-saves-millions-while-elevating-marketing-efforts>

Marketing Beat. (2024). Klarna marketing spend news. Retrieved from <https://www.marketing-beat.co.uk/2024/05/28/klarna-marketing-spend-news/>

Klarna. (2024a). Klarna announces profitable quarter as it gears up for biggest shopping season yet. Retrieved from <https://www.klarna.com/international/regulatory-news/klarna-announces-profitable-quarter-as-it-gears-up-for-biggest-shopping-season-yet/>

Klarna. (2024b). 90% of Klarna staff are using AI daily: Game-changer for productivity. Retrieved from <https://www.klarna.com/international/press/90-of-klarna-staff-are-using-ai-daily-game-changer-for-productivity/>

Klarna. (2024c). Klarna cashback drives $100M worth of sales in just six weeks. Retrieved from <https://www.klarna.com/international/press/klarna-cashback-drives-100m-worth-of-sales-in-just-six-weeks/>

Kshetri, N. (2024). Generative AI in Advertising. IT Professional, 26(5), 15–19. 10.1109/MITP.2024.3457328

Harvard Business School’s latest paper, Generative AI and the Nature of Work, has some eye-opening insights for creatives. Spoiler: AI won’t replace you just yet, but it might become your new admin assistant. Here’s what this could mean for us as designers—and why it might just make our lives a bit more interesting.

Less Admin, More Art

AI loves doing the boring bits. Just as developers can now focus on coding rather than email chains, designers might soon see AI handling all those tedious tasks like project scheduling and invoicing. This means more time for us to get back to the fun stuff—design, ideation, and actually creating.

A Solo Studio in Your Pocket

According to the study, generative AI fosters more autonomous work. So, instead of roping in team members every time you want feedback, you could use AI to create drafts, concept ideas, and generate a dozen logo options on the fly. You’ll still need to work with others, but AI is here to keep the solo creative company when everyone else is “in meetings.”

An Open Door for Experimentation

One of the perks of AI is its “let’s try anything” mentality. AI can generate endless variations without getting tired or bored, which means you can experiment with new design styles and techniques you wouldn’t usually risk in client work. Think of it as your creative lab partner who never judges your wildest ideas.

Skill Up or Ship Out

With AI picking up routine tasks, creatives will need to double down on their unique human skills—like big-picture thinking, storytelling, and empathy. So, while AI becomes a tool for doing, we’ll need to become experts in knowing what to do and why. This shift isn’t just about working faster; it’s about working smarter, too.

Collaborate Less, Create More

Finally, while AI might reduce the need for constant team check-ins, it could make the collaboration we do have more meaningful. With the basics handled by our digital co-workers, teams could meet less for the mundane and more for the bold, creative decisions that actually matter.

In Short

Generative AI won’t make creatives redundant—it might just make us better at what we actually want to do. So if you’re worried about AI “taking your job,” don’t be. It’s just taking the bits you probably weren’t that thrilled to do anyway.

Hoffmann, Manuel and Boysel, Sam and Nagle, Frank and Peng, Sida and Xu, Kevin, Generative AI and the Nature of Work (October 27, 2024). Harvard Business School Strategy Unit Working Paper No. 25-021, Harvard Business Working Paper No. No. 25-021, Available at SSRN: https://ssrn.com/abstract=5007084 or http://dx.doi.org/10.2139/ssrn.5007084

“What is a number, that a man may know it, and a man that he may know a number.”

In this rare 1962 interview filmed in Old Lyme, Connecticut, Warren McCulloch—barefoot, shirtless, and full of energy—feels like a mix of Aldous Huxley, George Orwell, and the actor Willem Dafoe, with his wild, piercing gaze. The conversation offers a glimpse into the mind of one of the early thinkers in artificial intelligence.

McCulloch’s work laid the foundation for many of the AI technologies that shape our world today, from neural networks and natural language processing to the swift advancements in generative AI.

His spooky words, “Don’t shake the table…” as the interview closes, echo with a certain finality—an early and important call for ethical consideration as AI progresses.