A philosophy called effective altruism has shaped today’s AI safety movement, and Shyam Sankar says Americans should think twice before trusting it.
Sankar, the chief technology officer and executive vice president at Palantir, argues that the latest wave of warnings about AI is not just a technical debate. It is a fight over control of the most important technology of the 21st century.
The immediate spark was a burst of public alarm last week. Jacob Coxon, a software engineer who resigned from Anthropic, said on X that 'the people building AI earnestly believe that it could kill us all by the end of the decade.' A current employee at a frontier lab then said he and many of his co-workers believe there is a significant chance AI will kill us all, putting the odds at perhaps greater than 10 percent.
Those claims set off calls for AI safety regulation from the leaders of almost every major AI company. But Sankar says the warnings should not be taken at face value.
He says they are infused with effective altruism, a philosophy that became familiar to many when crypto entrepreneur Sam Bankman-Fried said it motivated him. Bankman-Fried once presented himself as the 'good billionaire' who would make a fortune with FTX and give it away to protect humanity’s future, especially from big, existential risks. Instead, he stole billions from his customers and was sentenced to 25 years in prison.
That history matters because it exposes the danger of granting moral authority to a movement that presents itself as uniquely equipped to decide what risks society should tolerate. The record is plain: once concentrated power is handed to self-anointed guardians of the future, the public often discovers that the guardians want more power than accountability.
Follow the incentive, not the press release. When industry leaders and ideologues call for safety rules around a breakthrough technology, the first question is not whether they sound sober. It is who gains leverage over innovation, who pays the cost in slower progress, and whether ordinary Americans keep access to the tools that will define the next decade.
That is the deeper issue in Sankar’s warning. The debate over AI is being framed as a matter of existential caution, but the consequence may be far more familiar: more control for institutions, less freedom for markets, and fewer decisions left to the people who will live with the results.
What is true does not need an adjective. AI may carry real risks, and no serious country should ignore them. But a free society should also be wary when the language of moral urgency becomes a vehicle for bureaucratic control. The stakes are not only technological. They are political, economic, and ultimately about whether innovation remains open to the public or gets filtered through a new administrative state of experts.
