Several Silicon Valley billionaires backing the rapid expansion of artificial intelligence (AI) are also playing a significant role in shaping how the technology and its risks are perceived, industry sources say. This influence extends through funding research, testing labs, advocacy groups, and media coverage.

Central to this dynamic is Anthropic, an AI startup founded in 2021 that has swiftly risen to prominence with a valuation approaching $1 trillion. Key financial supporters include Dustin Moskovitz, a Facebook co-founder, and Sam Bankman-Fried, the now-incarcerated founder of cryptocurrency exchange FTX. Both figures have invested heavily not only in Anthropic but also in affiliated organizations involved in AI safety and journalism.

The backing is closely linked to the Effective Altruism (EA) movement, an ideological group focused on using evidence and reason to benefit society. According to insiders, this network has cultivated a powerful ecosystem that channels resources into Anthropic while reinforcing a message centered on AI risk and regulation.

Critics argue that this funding structure has contributed to an echo chamber that amplifies Anthropic’s perspective, particularly the cautious or “doomerist” stance articulated by CEO Dario Amodei. Some industry experts suggest this approach may be a strategic effort to influence the pace and direction of AI development.

“Nathan Zhao, founding partner at the tech investment firm SevenUp Capital, described a “cult-like” atmosphere in Anthropic’s early days, with employees demonstrating near-unquestioning loyalty to Amodei,” according to accounts from several current and former workers. Zhao also contended that Amodei’s recent calls for government regulation of AI primarily serve Anthropic’s interests, as smaller competitors might lack the capital to comply, potentially diminishing rival firms’ market presence.

Such sentiments have sparked discontent within the AI community. Richard Chang, an AI engineer who requested anonymity, said that while Anthropic once enjoyed widespread goodwill, industry views have soured amid what some perceive as high-handed posturing. “They always claim the moral high ground and criticize everyone else,” Chang said. He added that the message promulgated within Anthropic’s network frames the company as uniquely trustworthy, a notion that is aggressively promoted through various channels.

Financial connections extend beyond direct investment. Moskovitz is a board member of Coefficient Giving—a nonprofit formerly called Open Philanthropy—that supports the Tarbell Center for AI Journalism. The center funds more than 80 reporters embedded in influential outlets like Time, Bloomberg, The Guardian, and The Verge. The center emphasizes the editorial independence of its journalists.

Another nonprofit aligned with EA, Longview Philanthropy, finances Model Evaluation and Threat Research (METR), a company conducting benchmarking studies on AI systems. METR’s board shows close ties to both Anthropic and other industry players such as OpenAI.

Anthropic also operates programs to train and employ advocates extensively. Its “Claude Corps” recruits approximately 1,000 individuals to promote its chatbot Claude across over 400 nonprofit organizations worldwide. In addition, the company pays ambassadors and community leaders tasked with advancing its technology and safety narrative.

Some observers note a shift in how these efforts are framed. Brian Chau, a machine learning engineer and founder of an AI news site, explained that prior to Bankman-Fried's arrest in 2022, many such initiatives openly aligned with Effective Altruism principles. Since then, the emphasis has shifted more explicitly toward “AI safety,” with the movement’s label being downplayed, likely due to Bankman-Fried’s legal troubles after allegations of defrauding investors for billions of dollars.

Bankman-Fried led a $580 million Anthropic funding round in April 2022, months before his cryptocurrency enterprise collapsed. Federal authorities compelled the forced sale of his stake in the company as part of ongoing prosecutions.

Chau warned that the financial incentives within this ecosystem might cloud the judgment of those involved. “The more the incentives make it so you benefit from a certain belief, the less aware you are that’s why you believe it,” he said. He suggested this dynamic applies both to Amodei and many in the Effective Altruism movement, indicating potential conflicts between altruistic goals and business interests in the AI sector.