Human in the Loop

In October 2025, when Microsoft announced its restructured partnership with OpenAI, the numbers told a peculiar story. Microsoft now holds an investment valued at approximately $135 billion in OpenAI, representing roughly 27 per cent of the company. Meanwhile, OpenAI has contracted to purchase an incremental $250 billion of Azure services. The money flows in a perfect circle: investment becomes infrastructure spending becomes revenue becomes valuation becomes more investment. It's elegant, mathematically coherent, and possibly the blueprint for how artificial intelligence will either democratise intelligence or concentrate it in ways that make previous tech monopolies look quaint.

This isn't an isolated peculiarity. Amazon invested $8 billion in Anthropic throughout 2024, with the stipulation that Anthropic use Amazon's custom Trainium chips and AWS as its primary cloud provider. The investment returns to Amazon as infrastructure spending, counted as revenue, justifying more investment. When CoreWeave, the GPU cloud provider that went all-in on Nvidia, secured a $7.5 billion debt financing facility, Microsoft became its largest customer, accounting for 62 per cent of all revenue. Nvidia, meanwhile, holds approximately 5 per cent equity in CoreWeave, one of its largest chip customers.

The pattern repeats across the industry with mechanical precision. Major AI companies have engineered closed-loop financial ecosystems where investment, infrastructure ownership, and demand circulate among the same dominant players. The roles of customer, supplier, and investor have blurred into an indistinguishable whole. And while each deal, examined individually, makes perfect strategic sense, the cumulative effect raises questions that go beyond competition policy into something more fundamental: when organic growth becomes structurally indistinguishable from circular capital flows, how do we measure genuine market validation, and at what point does strategic vertical integration transition from competitive advantage to barriers that fundamentally reshape who gets to participate in building the AI-powered future?

The Architecture of Circularity

To understand how we arrived at this moment, you have to appreciate the sheer capital intensity of frontier AI development. When Meta released its Llama 3.1 model in 2024, estimates placed the development cost at approximately $170 million, excluding data acquisition and labour. That's just one model, from one company. Meta announced plans to expand its AI infrastructure to compute power equivalent to 600,000 Nvidia H100 GPUs by the end of 2024, representing an $18 billion investment in chips alone.

Across the industry, the four largest U.S. tech firms, Alphabet, Amazon, Meta, and Microsoft, collectively planned roughly $315 billion in capital spending for 2025, primarily on AI and cloud infrastructure. Capital spending by the top five U.S. hyperscalers rose 66 per cent to $211 billion in 2024. The numbers are staggering, but they reveal something crucial: the entry price for playing at the frontier of AI development has reached levels that exclude all but the largest, most capitalised organisations.

This capital intensity creates what economists call “natural” vertical integration, though there's nothing particularly natural about it. When you need tens of billions of pounds in infrastructure to train state-of-the-art models, and only a handful of companies possess both that infrastructure and the capital to build more, vertical integration isn't a strategic choice. It's gravity. Google's tight integration of foundation models across its entire stack, from custom TPU chips through Google Cloud to consumer products, represents this logic taken to its extreme. As industry analysts have noted, Google's vertical integration of AI functions similarly to Oracle's historical advantage from integrating software with hardware, a strategic moat competitors found nearly impossible to cross.

But what distinguishes the current moment from previous waves of tech consolidation is the recursive nature of the value flows. In traditional vertical integration, a company like Ford owned the mines that produced iron ore, the foundries that turned it into steel, the factories that assembled cars, and the dealerships that sold them. Value flowed in one direction: from raw materials to finished product to customer. The money ultimately came from outside the system.

In AI's circular economy, the money rarely leaves the system at all. Microsoft invests $13 billion in OpenAI. OpenAI commits to $250 billion in Azure spending. Microsoft records this as cloud revenue, which increases Azure's growth metrics, which justifies Microsoft's valuation, which enables more investment. But here's the critical detail: Microsoft recorded a $683 million expense related to its share of OpenAI's losses in Q1 fiscal 2025, with CFO Amy Hood expecting that figure to expand to $1.5 billion in Q2. The investment generates losses, which generate infrastructure spending, which generates revenue, which absorbs the losses. Whether end customers, the actual source of revenue outside this closed loop, are materialising in sufficient numbers to justify the cycle becomes surprisingly difficult to answer.

The Validation Problem

This creates what we might call the validation problem: how do you distinguish genuine market traction from structurally sustained momentum within self-reinforcing networks? OpenAI's 2025 revenue hit $12.7 billion, doubling from 2024. That's impressive growth by any standard. But as the exclusive provider of cloud computing services to OpenAI, Azure monetises all workloads involving OpenAI's large language models because they run on Microsoft's infrastructure. Microsoft's AI business is on pace to exceed a $10 billion annual revenue run rate, which the company claims “will be the fastest business in our history to reach this milestone.” But when your customer is also your investment, and their spending is your revenue, the traditional signals of market validation begin to behave strangely.

Wall Street analysts have become increasingly vocal about these concerns. Following the announcement of several high-profile circular deals in 2024, analysts raised questions about whether demand for AI could be overstated. As one industry observer noted, “There is a risk that money flowing between AI companies is creating a mirage of growth.” The concern isn't that the technology lacks value, but that the current financial architecture makes it nearly impossible to separate signal from noise, genuine adoption from circular capital flows.

The FTC has taken notice. In January 2024, the agency issued compulsory orders to Alphabet, Amazon, Anthropic, Microsoft, and OpenAI, launching what FTC Chair Lina Khan described as a “market inquiry into the investments and partnerships being formed between AI developers and major cloud service providers.” The partnerships involved more than $20 billion in cumulative financial investment. When the FTC issued its staff report in January 2025, the findings painted a detailed picture: equity and revenue-sharing rights retained by cloud providers, consultation and control rights gained through investments, and exclusivity arrangements that tie AI developers to specific infrastructure providers.

The report identified several competition concerns. The partnerships may impact access to computing resources and engineering talent, increase switching costs for AI developers, and provide cloud service provider partners with access to sensitive technical and business information unavailable to others. What the report describes, in essence, is not just vertical integration but something closer to vertical entanglement: relationships so complex and mutually dependent that extricating one party from another would require unwinding not just contracts but the fundamental business model.

The Concentration Engine

This financial architecture doesn't just reflect market concentration; it actively produces it. The mechanism is straightforward: capital intensity creates barriers to entry, vertical integration increases switching costs, and circular investment flows obscure market signals that might otherwise redirect capital toward alternatives.

Consider the GPU shortage that has characterised AI development since the generative AI boom began. During an FTC Tech Summit discussion in January 2024, participants noted that the dominance of big tech in cloud computing, coupled with a shortage of chips, was preventing smaller AI software and hardware startups from competing fairly. The major cloud providers control an estimated 66 per cent of the cloud computing market and have sway over who gets GPUs to train and run models.

A 2024 Stanford survey found that 67 per cent of AI startups couldn't access enough GPUs, forcing them to use slower CPUs or pay exorbitant cloud rates exceeding $3 per hour for an A100 GPU. The inflated costs and prolonged waiting times create significant economic barriers. Nvidia's V100 card costs over $10,000, with waiting periods surging to six months from order.

But here's where circular investment amplifies the concentration effect: when cloud providers invest in their customers, they simultaneously secure future demand for their infrastructure and gain insight into which startups might become competitive threats. Amazon's $8 billion investment in Anthropic came with the requirement that Anthropic use AWS as its primary cloud provider and train its models on Amazon's custom Trainium chips. Anthropic's models will scale to use more than 1 million of Amazon's Trainium2 chips for training and inference in 2025. This isn't just securing a customer; it's architecting the customer's technological dependencies.

The competitive dynamics this creates are subtle but profound. If you're a promising AI startup, you face a choice: accept investment and infrastructure support from a hyperscaler, which accelerates your development but ties your architecture to their ecosystem, or maintain independence but face potentially insurmountable resource constraints. Most choose the former. And with each choice, the circular economy grows denser, more interconnected, more difficult to penetrate from outside.

The data bears this out. In 2024, over 50 per cent of all global venture capital funding went to AI startups, totalling $131.5 billion, marking a 52 per cent year-over-year increase. Yet increasing infrastructure costs are raising barriers that, for some AI startups, may be insurmountable despite large fundraising rounds. Organisations boosted their spending on compute and storage hardware for AI deployments by 97 per cent year-over-year in the first half of 2024, totalling $47.4 billion. The capital flows primarily to companies that can either afford frontier-scale infrastructure or accept deep integration with those who can.

Innovation at the Edges

This raises perhaps the most consequential question: what happens to innovation velocity when the market concentrates in this way? The conventional wisdom in tech policy holds that competition drives innovation, that a diversity of approaches produces better outcomes. But AI appears to present a paradox: the capital requirements for frontier development seem to necessitate concentration, yet concentration risks exactly the kind of innovation stagnation that capital requirements were meant to prevent.

The evidence on innovation velocity is mixed and contested. Research measuring AI innovation pace found that in 2019, more than three AI preprints were submitted to arXiv per hour, over 148 times faster than in 1994. One deep learning-related preprint was submitted every 0.87 hours, over 1,064 times faster than in 1994. By these measures, AI innovation has never been faster. But these metrics measure quantity, not the diversity of approaches or the distribution of who gets to innovate.

BCG research in 2024 identified fintech, software, and banking as the sectors with the highest concentration of AI leaders, noting that AI-powered growth concentrates among larger firms and is associated with higher industry concentration. Other research found that firms with rich data resources can leverage large databases to reduce computational costs of training models and increase predictive accuracy, meaning organisations with bigger datasets have lower costs and better returns in AI production.

Yet dismissing the possibility of innovation outside these walled gardens would be premature. Meta's open-source Llama strategy represents a fascinating counterpoint to the closed, circular model dominating elsewhere. Since its release, Llama has seen more than 650 million downloads, averaging one million downloads per day since February 2023, making it the most adopted AI model. Meta's rationale for open-sourcing is revealing: since selling access to AI models isn't their business model, openly releasing Llama doesn't undercut their revenue the way it does for closed providers. More strategically, Meta shifts infrastructure costs outward. Developers using Llama models handle their own deployment and infrastructure, making Meta's approach capital efficient.

Mark Zuckerberg explicitly told investors that open-sourcing Llama is “not entirely altruistic,” that it will save Meta money. But the effect, intentional or not, is to create pathways for participation outside the circular economy. A researcher in Lagos, a startup in Jakarta, or a university lab in São Paulo can download Llama, fine-tune it for their specific needs, and deploy applications without accepting investment from, or owing infrastructure spending to, any hyperscaler.

The question is whether open-source models can keep pace with frontier development. The estimated cost of Llama 3.1, at $170 million excluding other expenses, suggests that even Meta's largesse has limits. If the performance gap between open and closed models widens beyond a certain threshold, open-source becomes a sandbox for experimentation rather than a genuine alternative for frontier applications. And if that happens, the circular economy becomes not just dominant but definitional.

The Global Dimension

These dynamics take on additional complexity when viewed through a global lens. As AI capabilities become increasingly central to economic competitiveness and national security, governments worldwide are grappling with questions of “sovereign AI,” the idea that nations need indigenous AI capabilities not wholly dependent on foreign infrastructure and models.

The UK's Department for Science, Innovation and Technology established the Sovereign AI Unit with up to £500 million in funding. Prime Minister Keir Starmer announced at London Tech Week a £2 billion commitment, with £1 billion towards AI-related investments, including new data centres. Data centres were classified as critical national infrastructure in September 2024. Nvidia responded by establishing the UK Sovereign AI Industry Forum, uniting leading UK businesses including Babcock, BAE Systems, Barclays, BT, National Grid, and Standard Chartered to advance sovereign AI infrastructure.

The EU has been more ambitious still. The €200 billion AI Continent Action Plan aims to establish European digital sovereignty and transform the EU into a global AI leader. The InvestAI programme promotes a “European preference” in public procurement for critical technologies, including AI chips and cloud infrastructure. London-based hyperscaler Nscale raised €936 million in Europe's largest Series B funding round to accelerate European sovereign AI infrastructure deployment.

But here's the paradox: building sovereign AI infrastructure requires exactly the kind of capital-intensive vertical integration that creates circular economies. The UK's partnership with Nvidia, the EU's preference for European providers, these aren't alternatives to the circular model. They're attempts to create national or regional versions of it. The structural logic they've pioneered, circular investment flows, vertical integration, infrastructure lock-in, appears to be the only economically viable path to frontier AI capabilities.

This creates a coordination problem at the global level. If every major economy pursues sovereign AI through vertically integrated national champions, we may end up with a fragmented landscape where models, infrastructure, and data pools don't interoperate, where switching costs between ecosystems become prohibitive. The alternative, accepting dependence on a handful of U.S.-based platforms, raises its own concerns about economic security, data sovereignty, and geopolitical leverage.

The developing world faces even more acute challenges. AI technology may lower barriers to entry for potential startup founders around the world, but investors remain unconvinced it will lead to increased activity in emerging markets. As one venture capitalist noted, “AI doesn't solve structural challenges faced by emerging markets,” pointing to limited funding availability, inadequate infrastructure, and challenges securing revenue. While AI funding exploded to more than $100 billion in 2024, up 80 per cent from 2023, this was heavily concentrated in established tech hubs rather than emerging markets.

The capital intensity barrier that affects startups in London or Berlin becomes insurmountable for entrepreneurs in Lagos or Dhaka. And because the circular economy concentrates not just capital but data, talent, and institutional knowledge within its loops, the gap between participants and non-participants widens with each investment cycle. The promise of AI democratising intelligence confronts the reality of an economic architecture that systematically excludes most of the world's population from meaningful participation.

Systemic Fragility

The circular economy also creates systemic risks that only become visible when you examine the network as a whole. Financial regulators have begun sounding warnings that echo, perhaps ominously, the concerns raised before previous bubbles burst.

In a 2024 analysis of AI in financial markets, regulators warned that widespread adoption of advanced AI models could heighten systemic risks and introduce novel forms of market manipulation. The concern centres on what researchers call “risk monoculture”: if multiple financial institutions rely on the same AI engine, it drives them to similar beliefs and actions, harmonising trading activities in ways that amplify procyclicality and create more booms and busts. Worse, if authorities also depend on the same AI engine for analytics, they may not be able to identify resulting fragilities until it's too late.

The parallel to AI infrastructure is uncomfortable but apt. If a small number of cloud providers supply the compute for a large fraction of AI development, if those same providers invest in their customers, if the customers' spending constitutes a significant fraction of the providers' revenue, then the whole system becomes vulnerable to correlated failures. A security breach affecting one major cloud provider could cascade across dozens of AI companies simultaneously. A miscalculation in one major investment could trigger a broader reassessment of valuations.

The Department of Homeland Security, in reports published throughout 2024, warned that deploying AI may make critical infrastructure systems supporting the nation's essential functions more vulnerable. While AI can present transformative solutions for critical infrastructure, it also carries the risk of making those systems vulnerable in new ways to critical failures, physical attacks, and cyber attacks.

CoreWeave illustrates these interdependencies in microcosm. The Nvidia-backed GPU cloud provider went from cryptocurrency mining to a $19 billion valuation based primarily on AI infrastructure offerings. The company reported revenue surging to $1.9 billion in 2024, a 737 per cent increase from the previous year. But its net loss also widened, reaching $863.4 million in 2024. With Microsoft accounting for 62 per cent of revenue and Nvidia holding 5 per cent equity while being CoreWeave's primary supplier, if any link in that chain weakens, Microsoft's demand, Nvidia's supply, CoreWeave's ability to service its $7.5 billion debt, the reverberations could extend far beyond one company.

Industry observers have drawn explicit comparisons to dot-com bubble patterns. One analysis warned that “a weak link could threaten the viability of the whole industry.” The concern isn't that AI lacks real applications or genuine value. The concern is that the circular financial architecture has decoupled short-term valuations and revenue metrics from the underlying pace of genuine adoption, creating conditions where the system could continue expanding long past the point where fundamentals would otherwise suggest caution.

Alternative Architectures

Given these challenges, it's worth asking whether alternative architectures exist, whether the circular economy is inevitable or whether we're simply in an early stage where other models haven't yet matured.

Decentralised AI infrastructure represents one potential alternative. According to PitchBook, investors deployed $436 million in decentralised AI in 2024, representing nearly 200 per cent growth compared to 2023. Projects like Bittensor, Ocean Protocol, and Akash Network aim to create infrastructure that doesn't depend on hyperscaler control. Akash Network, for instance, offers a decentralised compute marketplace with blockchain-based resource allocation for transparency and competitive pricing. Federated learning allows AI models to train on data while it remains locally stored, preserving privacy.

These approaches are promising but face substantial obstacles. Decentralised infrastructure still requires significant technical expertise. The performance and reliability of distributed systems often lag behind centralised hyperscaler offerings, particularly for the demanding workloads of frontier model training. And most fundamentally, decentralised approaches struggle with the cold-start problem: how do you bootstrap a network large enough to be useful when most developers already depend on established platforms?

Some AI companies are deliberately avoiding deep entanglements with cloud providers, maintaining multi-cloud strategies or building their own infrastructure. OpenAI's $300 billion cloud contract with Oracle starting in 2027 and partnerships with SoftBank on data centre projects represent attempts to reduce dependence on Microsoft's infrastructure, though these simply substitute one set of dependencies for others.

Regulatory intervention could reshape the landscape. The FTC's investigation, the EU's antitrust scrutiny, the Department of Justice's examination of Nvidia's practices, all suggest authorities recognise the competition concerns these circular relationships raise. In July 2024, the DOJ, FTC, UK Competition and Markets Authority, and European Commission released a joint statement specifying three concerns: concentrated control of key inputs, the ability of large incumbent digital firms to entrench or extend power in AI-related markets, and arrangements among key players that might reduce competition.

Specific investigations have targeted practices at the heart of the circular economy. The DOJ investigated whether Nvidia made it difficult for buyers to switch suppliers and penalised those that don't exclusively use its AI chips. The FTC sought information about Microsoft's partnership with OpenAI and whether it imposed licensing terms preventing customers from moving their data from Azure to competitors' services.

Yet regulatory intervention faces its own challenges. The global nature of AI development means that overly aggressive regulation in one jurisdiction might simply shift activity elsewhere. The complexity of these relationships makes it difficult to determine which arrangements enhance efficiency and which harm competition. And the speed of AI development creates a timing problem: by the time regulators fully understand one market structure, the industry may have evolved to another.

The Participation Question

Which brings us back to the fundamental question: at what point does strategic vertical integration transition from competitive advantage to barriers that fundamentally reshape who gets to participate in building the AI-powered future?

The data on participation is stark. While 40 per cent of small businesses reported some level of AI use in a 2024 McKinsey report, representing a 25 per cent increase in AI adoption over three years, the nature of that participation matters. Using AI tools is different from building them. Deploying models is different from training them. Being a customer in someone else's circular economy is different from being a participant in shaping what gets built.

Four common barriers block AI adoption for all companies: people, control of AI models, quality, and cost. Executives estimate that 40 per cent of their workforce will need reskilling in the next three years. Many talented innovators are unable to design, create, or own new AI models simply because they lack access to the computational infrastructure required to develop them. Even among companies adopting AI, 74 per cent struggle to achieve and scale value according to BCG research in 2024.

The concentration of AI capabilities within circular ecosystems doesn't just affect who builds models; it shapes what problems AI addresses. When development concentrates in Silicon Valley, Redmond, and Mountain View, funded by hyperscaler investment, deployed on hyperscaler infrastructure, the priorities reflect those environments. Applications that serve Western, English-speaking, affluent users receive disproportionate attention. Problems facing the global majority, from agricultural optimisation in smallholder farming to healthcare diagnostics in resource-constrained settings, receive less focus not because they're less important but because they're outside the incentive structures of circular capital flows.

This creates what we might call the representation problem: if the economic architecture of AI systematically excludes most of the world's population from meaningful participation in development, then AI capabilities, however powerful, will reflect the priorities, biases, and blind spots of the narrow slice of humanity that does participate. The promise of artificial general intelligence, assuming we ever achieve it, becomes the reality of narrow intelligence reflecting narrow interests.

Measuring What Matters

So how do we measure genuine market validation versus circular capital flows? How do we distinguish organic growth from structurally sustained momentum? The traditional metrics, revenue growth, customer acquisition, market share, all behave strangely in circular economies. When your investor is your customer and your customer is your revenue, the signals that normally guide capital allocation become noise.

We need new metrics, new frameworks for understanding what constitutes genuine traction in markets characterised by this degree of vertical integration and circular investment. Some possibilities suggest themselves. The diversity of revenue sources: how much of a company's revenue comes from entities that have also invested in it? The sustainability of unit economics: if circular investment stopped tomorrow, would the business model still work? The breadth of capability access: how many organisations, across how many geographies and economic strata, can actually utilise the technology being developed?

None of these are perfect, and all face measurement challenges. But the alternative, continuing to rely on metrics designed for different market structures, risks mistaking financial engineering for value creation until the distinction becomes a crisis.

The industry's response to these questions will shape not just competitive dynamics but the fundamental trajectory of artificial intelligence as a technology. If we accept that frontier AI development necessarily requires circular investment flows, that vertical integration is simply the efficient market structure for this technology, then we're also accepting that participation in AI's future belongs primarily to those already inside the loop.

If, alternatively, we view the current architecture as a contingent outcome of particular market conditions rather than inevitable necessity, then alternatives become worth pursuing. Open-source models like Llama, decentralised infrastructure like Akash, regulatory interventions that reduce switching costs and increase interoperability, sovereign AI initiatives that create regional alternatives, all represent paths toward a more distributed future.

The stakes extend beyond economics into questions of power, governance, and what kind of future AI helps create. Technologies that concentrate capability also concentrate influence over how those capabilities get used. If a handful of companies, bound together in mutually reinforcing investment relationships, control the infrastructure on which AI depends, they also control, directly or indirectly, what AI can do and who can do it.

The circular economy of AI infrastructure isn't a market failure in the traditional sense. Each individual transaction makes rational sense. Each investment serves legitimate strategic purposes. Each infrastructure partnership solves real coordination problems. But the emergent properties of the system as a whole, the concentration it produces, the barriers it creates, the fragilities it introduces, these are features that only become visible when you examine the network rather than the nodes.

And that network, as it currently exists, is rewiring the future of innovation in ways we're only beginning to understand. The money loops back on itself, investment becomes revenue becomes valuation becomes more investment. The question is what happens when, inevitably, the music stops. What happens when external demand, the revenue that comes from outside the circular flow, proves insufficient to justify the valuations the circle has created? What happens when the structural interdependencies that make the system efficient in good times make it fragile when conditions change?

We may be about to find out. The AI infrastructure buildout of 2024 and 2025 represents one of the largest capital deployments in technological history. The circular economy that's financing it represents one of the most intricate webs of financial interdependence the industry has created. And the future of who gets to participate in building AI-powered technologies hangs in the balance.

The answer to whether this architecture produces genuine innovation or systemic fragility, whether it democratises intelligence or concentrates it, whether it opens pathways to participation or closes them, won't be found in any single transaction or partnership. It will emerge from the cumulative effect of thousands of investment decisions, infrastructure commitments, and strategic choices. We're watching, in real time, as the financial architecture of AI either enables the most transformative technology in human history or constrains it within the same patterns of concentration and control that have characterised previous technological revolutions.

The loop is closing. The question is whether there's still time to open it.

---

Sources and References

1. Microsoft and OpenAI partnership restructuring (October 2025): Microsoft Official Blog, CNBC, TIME
2. Amazon-Anthropic investment relationship ($8 billion): CNBC, TechCrunch, PYMNTS
3. CoreWeave-Nvidia partnership and Microsoft customer relationship: PR Newswire, CNBC, Data Center Frontier
4. Meta Llama infrastructure investment ($18 billion in chips, $38-40 billion total): Meta AI Blog, The Register
5. Capital spending by hyperscalers ($211 billion in 2024, $315 billion planned 2025): Data Centre Magazine, multiple financial sources
6. Llama 3.1 development cost estimate ($170 million): NBER Working Paper, industry analysis
7. FTC AI market investigation and report (January 2024-2025): FTC official press releases and staff report
8. GPU shortage and accessibility statistics: Stanford survey 2024, The Register, FTC Tech Summit
9. AI startup funding ($131.5 billion, 52% increase): Multiple VC reports, industry analysis
10. Open-source Llama adoption (650 million downloads): Meta official statements
11. UK Sovereign AI initiatives (£2 billion commitment): UK Government, Department for Science, Innovation and Technology
12. EU AI Continent Action Plan (€200 billion): European Commission, WILLIAM FRY analysis
13. Decentralised AI infrastructure investment ($436 million): PitchBook 2024
14. Systemic risk analysis: DHS reports 2024, financial market AI analysis
15. DOJ, FTC, CMA, European Commission joint statement (July 2024): Official regulatory sources

---

Tim Green UK-based Systems Theorist & Independent Technology Writer

Tim explores the intersections of artificial intelligence, decentralised cognition, and posthuman ethics. His work, published at smarterarticles.co.uk, challenges dominant narratives of technological progress while proposing interdisciplinary frameworks for collective intelligence and digital stewardship.

His writing has been featured on Ground News and shared by independent researchers across both academic and technological communities.

ORCID: 0009-0002-0156-9795 Email: tim@smarterarticles.co.uk

Discuss...