Tech Leaders Clash in San Francisco Over Apocalyptic AI Warnings and Market Hype

Silicon Valley’s top technology executives and venture capitalists gathered in San Francisco this week for Goldman Sachs' annual investor summit at the Palace Hotel, where conversations quickly shifted from financial returns to a fierce dispute over the existential threats posed by artificial intelligence. Rather than focusing exclusively on balance sheets and customer growth, attendees found the gathering dominated by the fallout from the abrupt departure of Anthropic researcher Jacob Coxon.

The 27-year-old researcher, who previously worked at OpenAI before moving to its primary competitor Anthropic, made headlines after publicly resigning and warning that leading engineers believe their creations could ultimately wipe out humanity. Coxon warned that those constructing advanced artificial intelligence systems are gambling with human lives, asserting that developers will soon produce superhuman platforms capable of hacking virtually any system.

His remarks were not an isolated incident. A succession of prominent employees have stepped down from both Anthropic and OpenAI in recent years citing safety apprehensions, and current personnel quickly backed Coxon's statements online. Evan Hubinger, an Anthropic team lead, posted on social media platform X that researchers genuinely fear AI could cause human extinction, stating his personal estimate of that catastrophe occurring within the coming decade is higher than 10 percent.

Although Coxon firmly insisted his statements were genuine concerns rather than a marketing stunt, a growing number of corporate executives and venture capitalists have voiced deep cynicism regarding the flurry of doomsday warnings. With both Anthropic and OpenAI reportedly preparing for potentially historic initial public offerings, critics across the tech industry suspect that emphasizing the extreme dangers of AI is a calculated strategy to signal the unmatched power of their models.

Financial stakes remain exceptionally high, with Anthropic securing a valuation of $965 billion (£713 billion) during its most recent round of funding earlier this year. Skeptics argue that presenting models as practically omnipotent entities capable of altering human history is one of the few ways companies can justify such astronomical valuations to prospective public market investors.

Among the most vocal critics at the conference was George Arison, the chief executive of the LGBTQ+ dating application Grindr. Speaking to the BBC, Arison characterized the apocalyptic statements coming from Coxon and his colleagues as evidence of a deeply concerning, anti-civilisational worldview taking root at Anthropic.

Arison confirmed that the alarmist rhetoric had direct commercial consequences, prompting him to instruct Grindr engineers to halt their use of Anthropic’s technology. He argued that corporate leaders have a fiduciary duty to their shareholders that prevents them from relying on a software vendor that issues such extreme and destabilizing public statements.

The dating app executive questioned the true motives behind the public alarms, noting that whether leadership earnestly holds these beliefs or not, the claims conveniently generate investor excitement. According to Arison, claiming that an AI system will soon conquer every industry and displace entire workforces serves as a powerful narrative to support massive valuations.

The pushback extended to the highest levels of the semiconductor industry. Nvidia chief executive Jensen Huang addressed Coxon’s warnings before attendees at the San Francisco gathering, with multiple audience members confirming that Huang directly dismissed the claims as untrue.

Huang has previously rejected catastrophic narratives surrounding artificial intelligence, publicly describing assertions that the technology will bring about the end of humanity as complete nonsense. While Nvidia has a substantial financial interest in sustaining market confidence given its position as the leading supplier of AI processors, Huang's reaction mirrors an expanding fatigue among tech founders regarding apocalyptic forecasts.

Other prominent industry figures joined the public criticism. Brad Gerstner, who leads the investment firm Altimeter Capital, shared images of Huang from the event while dismissing Coxon’s resignation statements as ridiculous hyperbole.

Similarly, Clement Delangue, the chief executive of the AI platform Hugging Face, questioned the weight given to individual engineers on existential matters. Delangue, whose company was recently targeted in a breach by OpenAI agents and subsequently agreed to be acquired by Nvidia, remarked on X that consulting a model researcher about human extinction risks was akin to asking an air conditioning technician about global climate change.

Beyond allegations of hype, some industry observers argue that extreme safety warnings could serve an anti-competitive purpose. Critics suggest that encouraging fears of catastrophic risk may prompt heavy government interventions that stifle emerging competitors, effectively securing a market duopoly for well-capitalized frontrunners Anthropic and OpenAI.

Anthropic chief executive Dario Amodei has himself attracted controversy after previously warning that AI could eliminate half of all entry-level white-collar roles and test the endurance of humanity. However, in an essay published on Saturday, Amodei called for comprehensive global regulation and a deliberate slowdown in model development, emphasizing that the risks associated with rapid progress remain serious.

The publication of Amodei’s policy proposal prompted a more measured tone from several earlier critics. Gerstner described the Anthropic CEO's framework as an important step forward in balancing safety with technological acceleration, while Delangue offered to assist in enacting the suggested solutions, highlighting the uneasy ongoing debate over AI governance in Silicon Valley.