President Trump branded Anthropic “woke” and “radical left” as he banned the AI giant from serving the federal government last week – and newly surfaced blog posts from the company’s in-house “philosopher” could provide fresh fodder for critics, The Post has learned.
In a fiery Truth Social post on Friday, Trump said Anthropic – a rival to Sam Altman’s OpenAI and Elon Musk’s xAI that raised $30 billion last month in a deal valuing the firm at a whopping $380 billion – has been banned from serving any federal agency including the Department of War.
“The Leftwing nut jobs at Anthropic have made a DISASTROUS MISTAKE trying to STRONG-ARM the Department of War, and force them to obey their Terms of Service instead of our Constitution,” Trump wrote.
That was after an increasingly heated back-and-forth between Anthropic CEO Dario Amodei and War Secretary Pete Hegseth over Anthropic’s “red lines” that forbid government use of its AI tools for mass surveillance of Americans or to power weapons capable of firing without human oversight.
According to insiders, the squabble over “red lines” highlighted a growing concern about Anthropic within the Trump administration – namely, that the political views of the company’s top executives skewed left to an alarming degree.
Amodei himself is a Democratic donor who once called Trump a “feudal warlord” in a now-deleted social media post expressing support for Kamala Harris ahead of the 2024 election. In 2018, he described Trump as a “serious and legitimate threat to the rule of law.”
Those concerns haven’t been helped by recently unearthed blog posts by Amanda Askell, a 37-year-old, Oxford-educated research scientist who joined Anthropic in 2021 after a three-year stint at OpenAI, according to the sources.
“If you were to be told that ritual cannibalism was practiced by your friends, you would presumably say ‘either don’t serve me human flesh for dinner, or I’m not coming to your house,’” Askell wrote in an oddball 2016 post that likened serving meat to serving human flesh as it promoted vegetarianism.
The previous year, Askell had ripped incarceration as a form of punishment, likening it to “flogging.”
“If anything, the suffering imposed by imprisonment seems comparable to the suffering imposed by fairly severe forms of corporal punishment,” she wrote.
An Anthropic spokesperson said Askell’s blog posts were intended as an academic exercise rather than an expression of her personal views. The posts were written before she joined Anthropic and are unrelated to her work at the firm, the spokesperson added.
Askell didn’t respond to requests for comment.
Askell’s posts were the latest in a slew of data points that have sparked concerns among US officials about Anthropic’s suitability as a government contractor, according to sources.
“Secretary Hegseth is rebuilding the military and removing woke influences, so it’s clear Askell and Anthropic’s ideology is at the heart of this latest showdown,” one Washington-based AI policy insider told The Post.
As The Post reported in November, the firm’s ties to Democratic megadonors like LinkedIn cofounder Reid Hoffman, as well as the cult-like “Effective Altruism” movement, have been “on the radar” of key decision makers within the Trump administration. Its backers also include the left-wing Ford Foundation, which has been singled out for criticism by Vice President JD Vance.
Hegseth specifically called out the ideology in a Friday post on X, writing: “Cloaked in the sanctimonious rhetoric of “effective altruism,” they have attempted to strong-arm the United States military into submission – a cowardly act of corporate virtue-signaling that places Silicon Valley ideology above American lives.”
Meanwhile, concerns about Askell were recently aired by Elon Musk, who has accused Anthropic of producing an AI model that “hates Whites & Asians, especially Chinese, heterosexuals and men.”
Musk claimed that Askell – who led the team responsible for creating an 80-page “soul document” that governs the Claude chatbot’s moral compass – isn’t qualified for the task because “Those without children lack a stake in the future.”
While Askell’s rambling posts frequently delve into heady apolitical philosophy, at least several signal strong progressive leanings.
“Getting rid of the unfairness that we have inherited from the past – such as different levels of investment in education and health across nations and social groups – may require proactive interventions,” Askell wrote in what looked like a pro-affirmative action post in 2020.
“We may even want to make decisions that are less procedurally fair in the short-term if doing so will reduce societal unfairness in long-term,” she added.
Askell was previously married to William MacAskill, who is seen as godfather of Effective Altruism, which was famously promoted by disgraced crypto tycoon Sam Bankman-Fried.
“Part of what attracted me to the effective altruism movement was the idea that I might be able to help people who are suffering from illnesses abroad,” Askell wrote in a 2017 blog post.
Nirit Weiss-Blatt, author of the “AI Panic” newsletter, said Anthropic is facing “backlash” for cultivating “fringe beliefs” and trying to take them “mainstream.”
“From its inception, Anthropic has marketed itself as ‘only we can tame the AI monster,’” Weiss-Blatt said. “The team’s mindset was shaped by the insular groupthink of the Effective Altruism movement.”
White House AI czar David Sacks is one of Anthropic’s most outspoken critics and has accused company brass of wanting to “backdoor Woke AI and other AI regulations through Blue states like California.”
Following a profile in the Wall Street Journal earlier this month, Askell posted on X: “A lot of the response has been people trying to infer my personal political views. For what it’s worth, I try to treat my personal political views as a potential source of bias and not as something it would be appropriate to try to train models to adopt.”
Weiss-Blatt argued that Anthropic has projected moral authority as a mechanism to “pursue regulatory capture, greater centralization, and control.”
“While the company warns of ‘power-seeking AI,’ it has itself become the power-seeking entity,” she added.
