Anthropic published a policy paper this spring arguing that American AI leadership is essential because the alternative is "authoritarian governments like the Chinese Communist Party" setting the terms for how AI develops globally. Two months later, the Emirati government's investment arm finished contributing to a $65 billion funding round that brought Anthropic's valuation to nearly $1 trillion. The Intercept, in a piece published Friday, puts both facts in the same sentence and asks what the combination means.
The tension between stated principles and actual financing is not unusual in the technology industry. What makes Anthropic's situation distinctive is that the company has built its public identity more explicitly around safety, ethics, and democratic values than any of its direct competitors. That positioning makes the distance between argument and practice more visible, and more consequential, as the company prepares for a public stock offering that will require institutional investors to assess exactly this kind of governance question.
Who Owns a Piece of Anthropic
Abu Dhabi's MGX, the AI-focused investment vehicle of the Emirati royal family, co-led Anthropic's $30 billion Series G in February 2026 at a $380 billion valuation. It returned to participate in the $65 billion Series H in May, which put the valuation at $965 billion. The chair of MGX is Sheikh Tahnoun bin Zayed Al Nahyan, Abu Dhabi's national security and intelligence chief. Sheikh Tahnoun is also the chairperson of G42, an Emirati AI company that, according to The Intercept, has already had reason to deploy Anthropic's large language models in its operations.
The United Arab Emirates does not permit organized political opposition, does not have an independent free press, and provides no meaningful protections for freedom of assembly or speech. Political dissidents have described torture in Emirati detention. Under Emirati law, speech that causes "damage to national unity" can result in life imprisonment, and capital punishment is available for some categories of the offense. These are not contested facts about UAE governance; they are documented in reports from Human Rights Watch, Amnesty International, and the US State Department's own annual human rights assessments.
Key Facts
- February 2026MGX co-leads Anthropic's $30B Series G at $380B valuation
- May 2026MGX participates in Anthropic's $65B Series H, valuing company at $965B
- MGX chairSheikh Tahnoun bin Zayed Al Nahyan, UAE national security and intelligence chief
- G42 connectionMGX-linked firm reportedly deployed Anthropic's frontier models
- Anthropic policy paperCalls for US to counter "authoritarian governments like the Chinese Communist Party"
- Dario Amodei memoAcknowledged Gulf investments enrich "dictators" but called it a "necessary compromise"
The Policy Paper and the Contradiction
Anthropic's policy paper, the basis of The Intercept's critique, argued that the United States and its democratic allies must dominate AI development because ceding ground would allow authoritarian states to export AI systems built on values antithetical to democracy. The argument is widely made in Washington policy circles, and versions of it appear in public statements from executives at OpenAI, Google DeepMind, and Meta AI as well. Anthropic's version was notable mainly for being more explicit about naming China as the threat.
The Intercept's piece observes that the UAE, which has invested in Anthropic across two consecutive funding rounds, operates an AI surveillance infrastructure that functions similarly to what the policy paper warns about. The critique is summarized in a passage The Intercept quotes directly: "The fact that Anthropic is partly owned by the government of Abu Dhabi, which is similar to China in its extensive use of AI surveillance to support an authoritarian government, suggests that its anti-authoritarian arguments are more based on a cynical policy position than a sincere passion for democracy."
"The fact that Anthropic is partly owned by the government of Abu Dhabi, which is similar to China in its extensive use of AI surveillance to support an authoritarian government, suggests that its anti-authoritarian arguments are more based on a cynical policy position than a sincere passion for democracy." Critic quoted in The Intercept, June 6, 2026
Amodei's Own Words
The Intercept's reporting draws on a memo from Dario Amodei that leaked earlier this year. In it, Amodei described accepting investment from state-backed Middle Eastern sovereign funds as a "necessary compromise." He acknowledged that the arrangement would, in his words, enrich "dictators," but framed it as unavoidable given the capital requirements of frontier AI development, which now run into the tens of billions per training run. The memo suggests Anthropic's leadership was aware of the tension long before the policy paper was published.
Anthropic has not responded publicly to The Intercept's piece. The company's IPO filings, expected to become public when the S-1 is declared effective by the SEC, will disclose more about the structure and terms of MGX's participation in both rounds. That will give potential public investors a clearer picture of how Anthropic plans to navigate its relationships with geopolitically sensitive capital as a listed company subject to shareholder scrutiny. The question of whether to treat Gulf sovereign investment as a material governance risk is one that institutional investors, particularly those with public mandates around democracy and human rights, will have to answer for themselves.
OpenAI has accepted investment from SoftBank, which manages Saudi capital through various vehicles. Microsoft and Google have extensive government relationships across the Gulf region. The difference, as The Intercept frames it, is the specificity of Anthropic's democratic branding against which the financing now sits. A company that campaigns against authoritarian AI in policy papers faces a different standard of scrutiny than one that makes no such claims.
