When does AI funding become control?

Introduction

Advanced AI may eventually help drive a much larger expansion of human capability than ordinary economic growth: accelerating science, extending healthy life, improving education, and making intelligence itself more widely available. But those possibilities depend not only on technical progress. They also depend on who controls the organisations building the most powerful systems.

One of the most important and least intuitive governance questions is whether frontier AI labs remain genuinely independent when they depend on a handful of technology giants for money, computing infrastructure and distribution. Many leading AI developers present themselves as separate companies with distinct missions. Yet some rely on cloud providers for billions of pounds in investment, access to specialised chips, training infrastructure and routes to market. Regulators increasingly worry that influence can emerge through partnerships long before a formal acquisition takes place. The central question is not whether cooperation should exist. It is when funding, infrastructure dependence and contractual rights become a form of hidden control.

Why frontier AI labs need cloud and capital partnerships

The largest frontier AI models require extraordinary levels of spending. Training advanced systems involves enormous data centres, specialised AI accelerators, engineering teams and electricity consumption. Only a small number of companies possess the infrastructure needed to support development at the frontier.

This has created a distinctive industry structure. Rather than buying AI labs outright, large technology firms often form strategic partnerships with them. The cloud company provides funding, computing resources and commercial distribution. The AI lab provides models, research talent and future technological capabilities.

The Microsoft–OpenAI relationship became the defining example. Microsoft invested billions of dollars into OpenAI while becoming its primary cloud provider and integrating OpenAI technology throughout Azure and Microsoft’s software ecosystem. Similar patterns emerged elsewhere. Amazon invested heavily in Anthropic while making Anthropic models available through Amazon Bedrock. Google invested in Anthropic and supplied cloud infrastructure while simultaneously competing against Anthropic through its own Gemini models.[Federal Trade Commission]ftc.govlaunches inquiry generative ai investments partnershipsMicrosoft and OpenAI, Amazon and Anthropic, and Google and Anthropic.Read more…[The Guardian]theguardian.comgoogle anthropic partnership cma aiUK regulator looks at Google's partnership with AnthropicJul 30, 2024 — The Competition and Markets Authority has begun a preliminary inv…

These arrangements solve real problems for frontier labs:

• They provide access to scarce computing power.

From one perspective, these deals can accelerate the development of AI systems that may eventually contribute to scientific discovery, medical research and broader forms of human flourishing. Without large-scale financing, some frontier research programmes might proceed far more slowly.

The concern is that dependence can gradually reshape incentives. If a lab relies on one infrastructure provider for most of its compute, switching becomes difficult even if no formal exclusivity exists.

Why regulators worry about influence without ownership

Traditional merger law was built around a relatively simple question: does one company acquire another?

Modern AI partnerships are more complicated. A cloud provider may not own a majority stake in an AI lab. It may not appoint a majority of directors. Yet it may still acquire significant influence through a combination of investment rights, infrastructure dependence, revenue-sharing arrangements, intellectual-property agreements and commercial integration.

This is why regulators increasingly discuss concepts such as “material influence”, “de facto control” and strategic dependency rather than focusing solely on ownership percentages.

The US Federal Trade Commission launched a major inquiry into partnerships involving Microsoft and OpenAI, Amazon and Anthropic, and Google and Anthropic. The FTC argued that cloud providers and AI developers were creating new forms of strategic relationships that could affect competition, access to computing resources and future market structure.[Heise]heise.deMicrosoft Google FTC investigates AI partnerships and collaborations 10247893Microsoft, Google: FTC investigates AI partnerships and…19 Jan 2025 — The US Federal Trade Commission (FTC) is concerned about collabo…[Federal Trade Commission]ftc.govlaunches inquiry generative ai investments partnershipsMicrosoft and OpenAI, Amazon and Anthropic, and Google and Anthropic.Read more…[Federal Trade Commission]ftc.govlaunches inquiry generative ai investments partnershipsMicrosoft and OpenAI, Amazon and Anthropic, and Google and Anthropic.Read more…

Several concerns appear repeatedly in regulatory reviews.

Compute dependence can create leverage

Frontier AI increasingly depends on specialised infrastructure that cannot easily be replicated.

A cloud provider that supplies most of an AI lab’s compute capacity may gain influence even without formal voting control. The lab may technically remain independent while facing enormous practical costs if it attempts to move elsewhere.

Regulators worry that dependence on a single infrastructure partner could affect strategic decisions, pricing choices or future partnerships. The concern is not necessarily overt interference. Influence can emerge because the infrastructure provider becomes difficult to replace.

Information access may alter competition

Cloud providers often host and support multiple AI developers simultaneously.

Competition authorities have raised concerns that infrastructure partnerships can create privileged access to commercially sensitive information. A company may gain insight into usage patterns, infrastructure requirements or technical development trajectories of firms that are nominally independent.[TechCrunch]techcrunch.comftc says partnerships like microsoft openai raise antitrust concernsFTC says partnerships like Microsoft-OpenAI raise antitrust…18 Jan 2025 — The report focuses on the roles of Microsoft, Amaz…

Even when strict safeguards exist, regulators worry that information asymmetries could strengthen already dominant firms.

Distribution agreements can reinforce concentration

Many cloud companies are not merely infrastructure suppliers. They also operate software platforms, enterprise ecosystems and AI marketplaces.

If a major cloud provider receives preferential rights to distribute or integrate frontier models, it can strengthen its position in multiple adjacent markets at once. This can make it harder for rivals to compete even if the AI lab itself remains formally separate.

The OpenAI debate: when does partnership become control?

No relationship has attracted more scrutiny than Microsoft’s partnership with OpenAI.

The scale of Microsoft’s investment, combined with OpenAI’s reliance on Azure infrastructure and Microsoft’s commercial rights, led regulators on both sides of the Atlantic to examine whether the relationship effectively resembled a merger without an acquisition. The UK’s Competition and Markets Authority opened a review specifically to examine whether developments in the partnership had created a relevant merger situation.[GOV.UK]GOV.UKcma seeks views on microsofts partnership with openaiCMA seeks views on Microsoft's partnership with OpenAIDec 8, 2023 — The CMA is now inviting views on whether the partnership between Micr…

The debate intensified after OpenAI’s board crisis in late 2023. Critics argued that the episode revealed how deeply intertwined the organisations had become. Supporters of the partnership argued the opposite: OpenAI’s board demonstrated independent authority by making major decisions without Microsoft’s control.

The CMA ultimately concluded that Microsoft did not exercise the level of control required for a merger investigation under UK law. Its final decision found that the partnership did not qualify for review under merger provisions.[GOV.UK]GOV.UKMicrosoft / Open AI partnership merger inquiry Found not to qualify decision5 March 2025: The CMA has decided that Microsoft's partnership with OpenAI does not qualify for investigation under the merger…Read more…Published: March 2025[GOV.UK]GOV.UKOpen source on gov.uk.

Yet the case remains important because it exposed a wider policy problem. A partnership can transform an industry’s structure even if it falls below existing legal thresholds for control.

The question is no longer simply whether a company owns an AI lab. It is whether modern competition frameworks adequately capture relationships built from infrastructure dependence, long-term contracts and strategic investment.

Anthropic and the rise of multi-partner dependence

Anthropic illustrates a different model.

Rather than aligning exclusively with one major technology company, Anthropic developed significant relationships with both Amazon and Google. Amazon committed billions of dollars in investment while becoming a major infrastructure partner. Google also invested heavily and supplied cloud services. The arrangements prompted scrutiny from UK and US regulators concerned about how strategic investments and cloud agreements could affect future competition.[GOV.UK]GOV.UKAlphabet IncGoogle LLC) / Anthropic merger inquiry19 November 2024: The CMA has decided that Alphabet's partnership with Anthropic does not qualify…Published: November 2024[GOV.UK]GOV.UKmicrosoft slash inflection ai inquiry/ Inflection inquiryPhase 1. CMA clearance decision. 4 September 2024: The CMA has cleared Microsoft's hiring of certain former employees…Published: September 2024

Regulators eventually concluded that neither the Amazon–Anthropic nor Google–Anthropic partnerships met the legal threshold for merger investigation under UK rules.[Practical Law]uk.practicallaw.thomsonreuters.comfull text decision finding that Amazon / Anthropic…CMA decides that Amazon / Anthropic partnership does not qualify for investigation…

Even so, the Anthropic case highlights a broader tension.

Multiple partnerships may reduce dependence on any single company. At the same time, they can deepen the overall integration of frontier AI labs into a small network of dominant infrastructure providers. If most frontier development depends on the same few cloud firms, competition may remain constrained even without exclusive relationships.

This matters because the long-term AI bloom vision depends partly on broad access to advanced intelligence. A world where many organisations can build, adapt and deploy powerful AI may produce different outcomes from a world where access flows through a small number of infrastructure gatekeepers.

The hidden governance question behind AI deals

The most important issue may not be competition in the narrow economic sense.

Frontier AI labs increasingly make decisions that affect public safety, scientific priorities, security practices and the direction of technological development. If future systems become substantially more capable, these choices could influence the trajectory of civilisation itself.

Partnerships therefore raise governance questions alongside antitrust questions.

For example:

Formal corporate structures do not always answer these questions clearly.

An organisation may be legally independent while depending on another company for financing, compute and market access. Influence can emerge through practical necessity rather than explicit authority.

This possibility explains why some researchers and policymakers describe AI partnerships as a new form of governance challenge rather than merely a competition issue.

How transparency and merger review could adapt

Regulators are not generally trying to ban partnerships. Most observers accept that frontier AI development currently requires large-scale infrastructure cooperation.

The harder question is how to make influence visible.

Several proposals have emerged.

Stronger disclosure requirements

Many AI partnerships involve complicated arrangements covering cloud commitments, licensing rights, revenue sharing and future investment options.

Greater disclosure could help regulators, researchers and the public understand how much independence frontier labs actually retain. The goal is not to expose trade secrets but to reveal governance-relevant relationships.

Reviewing influence rather than ownership alone

Traditional merger frameworks often focus on equity stakes and voting rights.

AI partnerships may require broader tests that examine:

Competition authorities in the UK, United States and elsewhere have increasingly explored these questions through partnership reviews and market studies.[Federal Trade Commission]ftc.govlaunches inquiry generative ai investments partnershipsMicrosoft and OpenAI, Amazon and Anthropic, and Google and Anthropic.Read more…[GOV.UK]GOV.UKcma seeks views on microsofts partnership with openaiCMA seeks views on Microsoft's partnership with OpenAIDec 8, 2023 — The CMA is now inviting views on whether the partnership between Micr…

Monitoring compute bottlenecks

Some analysts argue that oversight should focus less on ownership and more on access to computing resources.

If a small number of companies control the infrastructure required for frontier AI development, competition and innovation may become constrained regardless of who technically owns the leading labs.

This has encouraged growing interest in compute governance, cloud competition and public-interest access to advanced infrastructure.

Independent governance commitments

Another possibility is requiring clearer separation between investors and safety-critical decisions.

Some proposals include independent boards, transparent safety frameworks, external auditing and disclosure of conflicts of interest when major investors also act as infrastructure providers or commercial partners.

The goal is not to eliminate commercial incentives. It is to ensure that decisions with broad societal consequences are not shaped solely by private contractual relationships.

Why this matters for an AI-enabled human future

The optimistic case for AI assumes that increasingly powerful systems help expand human knowledge, productivity and freedom on a historically unprecedented scale. If advanced AI accelerates medicine, energy innovation, materials science and scientific discovery, the benefits could extend far beyond ordinary technology cycles.

But the distribution of those benefits depends partly on institutional structure.

A future where advanced intelligence is broadly accessible may look very different from one in which a handful of firms act as indispensable intermediaries between society and frontier AI capabilities.

The key governance challenge is therefore not whether partnerships should exist. Frontier AI almost certainly requires collaboration between researchers, investors and infrastructure providers. The challenge is ensuring that partnerships designed to fund innovation do not quietly become mechanisms through which strategic control over civilisation-shaping technologies becomes concentrated without meaningful public scrutiny.

As AI capabilities advance, questions about ownership, influence and dependency may become as important as questions about the technology itself. The future shape of an AI-enabled bloom could depend not only on how intelligent the systems become, but on who ultimately governs the institutions that build them.

Amazon book picks

Further Reading

Books and field guides related to When does AI funding become control?. Use these as the next step if you want deeper reading beyond the article.

Book

The Big Nine

By Amy Webb

Addresses concentration of influence among major AI organizations.

See on Amazon

Book

Power and Prediction

By Ajay Agrawal, Joshua Gans et al.

Explains incentives driving partnerships and investment in AI infrastructure.

See on Amazon

Book

The Master Switch

By Tim Wu

Offers historical parallels for control emerging through infrastructure dominance.

See on Amazon

Book

Competition Overdose

By Maurice E. Stucke, Ariel Ezrachi

Examines market power and regulatory concerns around concentration.

See on Amazon

Browse more on Amazon:The Big NinePower and PredictionThe Master Switch

As an Amazon Associate I earn from qualifying purchases.

eBay marketplace picks

Marketplace Samples

Example marketplace items related to this page. Use the search link to explore similar finds on eBay.

Shop location

UsingUSA USA USAUKCanadaAustraliaIreland

Example eBay listing

A.I. Artificial Intelligence Film Poster/Print A3/A4/A5 230gsm Framed

SearcheBay.co.uk: artificial intelligence poster

Browse similar oneBay.co.uk

Example eBay listing

Artificial intelligence is no a mat Framed Wall Art Poster Canvas Print Picture

SearcheBay.co.uk: artificial intelligence poster

Browse similar oneBay.co.uk

Example eBay listing

A.I. Artificial Intelligence Movie Film Poster Art Print

SearcheBay.co.uk: artificial intelligence poster

Browse similar oneBay.co.uk

Example eBay listing

A.I. - Artificial Intelligence Movie/Film Poster Art PICTURE / PRINT 9" x 8"

SearcheBay.co.uk: artificial intelligence poster

Browse similar oneBay.co.uk

Browse more oneBay.co.uk

Example items shown for inspiration; availability and pricing can change. Branchoria may earn a commission if you purchase through outbound eBay links.

Endnotes

1. Source: ftc.gov
Title: launches inquiry generative ai investments partnerships
Link:https://www.ftc.gov/news-events/news/press-releases/2024/01/ftc-launches-inquiry-generative-ai-investments-partnerships

2. Source: ftc.gov
Link:https://www.ftc.gov/system/files/ftc_gov/pdf/p246201_aipartnerships6breport_redacted_0.pdf

3. Source: techcrunch.com
Title: ftc says partnerships like microsoft openai raise antitrust concerns
Link:https://techcrunch.com/2025/01/18/ftc-says-partnerships-like-microsoft-openai-raise-antitrust-concerns/

4. Source: heise.de
Title: Microsoft Google FTC investigates AI partnerships and collaborations 10247893
Link:https://www.heise.de/en/news/Microsoft-Google-FTC-investigates-AI-partnerships-and-collaborations-10247893.html

5. Source: GOV.UK
Title: cma seeks views on microsofts partnership with openai
Link:https://www.gov.uk/government/news/cma-seeks-views-on-microsofts-partnership-with-openai

6. Source: GOV.UK
Title: Microsoft / Open AI partnership merger inquiry Found not to qualify decision
Link:https://www.gov.uk/cma-cases/microsoft-slash-openai-partnership-merger-inquiry

Published: March 2025

7. Source: assets.publishing.service.gov.uk
Link:https://assets.publishing.service.gov.uk/media/67c841d6d0fba2f1334cf276/1.MS.OAI-_Summary.pdf

8. Source: GOV.UK
Link:https://www.gov.uk/cma-cases/amazon-slash-anthropic-partnership-merger-inquiry

9. Source: GOV.UK
Title: Alphabet Inc
Link:https://www.gov.uk/cma-cases/alphabet-inc-google-llc-slash-anthropic-merger-inquiry

Published: November 2024

10. Source: assets.publishing.service.gov.uk
Title: Full text decision
Link:https://assets.publishing.service.gov.uk/media/6710ba44e84ae1fd8592f52c/Full_text_decision.pdf

Published: June 2024

11. Source: assets.publishing.service.gov.uk
Title: Full text decision
Link:https://assets.publishing.service.gov.uk/media/676959bae6ff7c8a1fde9d33/Full_text_decision__.pdf

Published: April 2024

12. Source: ftc.gov
Title: behind ftcs 6b report large ai partnerships investments
Link:https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/2025/01/behind-ftcs-6b-report-large-ai-partnerships-investments

13. Source: blogs.microsoft.com
Title: and openai joint statement on continuing partnership
Link:https://blogs.microsoft.com/blog/2026/02/27/microsoft-and-openai-joint-statement-on-continuing-partnership/

14. Source: GOV.UK
Title: microsoft slash inflection ai inquiry
Link:https://www.gov.uk/cma-cases/microsoft-slash-inflection-ai-inquiry

Published: September 2024

15. Source: theguardian.com
Title: google anthropic partnership cma ai
Link:https://www.theguardian.com/technology/article/2024/jul/30/google-anthropic-partnership-cma-ai

16. Source: theguardian.com
Title: anthropic ai startup uk regulator cma amazon investment
Link:https://www.theguardian.com/technology/article/2024/aug/08/anthropic-ai-startup-uk-regulator-cma-amazon-investment

17. Source: theguardian.com
Link:https://www.theguardian.com/technology/2025/mar/05/uk-cma-drops-competition-review-microsoft-openai-partnership

18. Source: uk.practicallaw.thomsonreuters.com
Link:https://uk.practicallaw.thomsonreuters.com/w-044-7449?contextData=%28sc.Default%29&transitionType=Default

19. Source: theguardian.com
Title: ftc ai inquiry microsoft alphabet amazon
Link:https://www.theguardian.com/technology/2024/jan/25/ftc-ai-inquiry-microsoft-alphabet-amazon

20. Source: concurrences.com
Title: Part VIII
Link:https://www.concurrences.com/en/review/issues/no-8-2025/on-topic/cloud-and-competition-policy/cloud-and-competition-policy-part-viii-cloud-service-ai-partnerships-the-ftc-s

21. Source: troutman.com
Link:https://www.troutman.com/insights/cloud-and-competition-policy-part-viii-cloud-service-ai-partnerships-the-ftcs-section-6b-report-and-its-antitrust-implications-in-the-trump-20-administration/

Additional References

22. Source: cma-cgm.com
Link:https://www.cma-cgm.com/

23. Source: linkedin.com
Link:https://www.linkedin.com/posts/joel-bamford-76593656_amazon-anthropic-partnership-merger-inquiry-activity-7245389776944529410-s1A6

24. Source: linkedin.com
Link:https://www.linkedin.com/pulse/what-key-takeaways-from-our-review-microsoftopenai-joel-bamford-auzre

25. Source: aifray.com
Link:https://aifray.com/ftc-releases-staff-report-on-ai-partnerships-investments-collection-of-information-and-innuendo-with-unclear-contours/

26. Source: jsm.com
Link:https://www.jsm.com/publications/2025/when-ai-meets-antitrust-how-a-strategic-ai-partnership-sparked-merger-control-scrutiny/

Published: May 2025

27. Source: datacenterdynamics.com
Title: cma launches full scale investigation into amazons partnership with anthropic
Link:https://www.datacenterdynamics.com/en/news/cma-launches-full-scale-investigation-into-amazons-partnership-with-anthropic/

28. Source: news.sky.com
Title: amazons 3bn ai anthropic investment investigated by uk regulator 13193450
Link:https://news.sky.com/story/amazons-3bn-ai-anthropic-investment-investigated-by-uk-regulator-13193450

29. Source: inquisitiveminds.bristows.com
Title: google anthropic the latest chapter in the cmas scrutiny of generative ai
Link:https://inquisitiveminds.bristows.com/post/102jnrs/google-anthropic-the-latest-chapter-in-the-cmas-scrutiny-of-generative-ai

30. Source: techmonitor.ai
Title: cma dismisses investigation into alphabet deal with ai firm anthropic
Link:https://www.techmonitor.ai/ai-and-automation/cma-dismisses-investigation-into-alphabet-deal-with-ai-firm-anthropic/

31. Source: reuters.com
Title: democratic us senators question google microsofts ai deals 2025 04 08
Link:https://www.reuters.com/sustainability/boards-policy-regulation/democratic-us-senators-question-google-microsofts-ai-deals-2025-04-08/