Internet sleuths have dug up old posts that appear to belong to Jacob Coxon, the former Anthropic employee who kicked the AI safety movement into overdrive with his sensational claim that the tech “could kill us all by the end of the decade.”
Coxon’s current X account, @hilbertspaess, was created in 2026, but the gumshoes found that there were two previous accounts on that platform under the same handle between 2021 and 2024. The first account has undergone several different display names. But its first name, as far back as November 2021, appears to be “Jacob Coxon,” <a href=”https://web.archive.org/web/20211122191701/
“>archived
data shows. The evidence linking the second account made in 2023 to Coxon is more tenuous, but appear to be written in the same style and cover the same topics as the first. (One of the detectives compiled the tweets into an easily searchable database here.)
Given Coxon’s drum-beating about AI’s supposed existential risks to humankind, observers have paid particular attention to the account’s posts on climate change, something that poses a definite existential risk — not just to humans but all life on the planet — and, of course, is scientifically verified and already underway.
Tell that to Coxon, though.
“it’s literally 2020 already. look at climate change headlines from the 1970s then look around. nothing is happening, use your eyes,” the account <a href=”https://web.archive.org/web/20220504234244/
“>wrote
in March 2022.
People who are “worried about climate change,” the account <a href=”https://web.archive.org/web/20220504233753/
“>opined
, are “beyond reason.” Perhaps if the planet gets hotter, there might be “positive consequences.”
“why it all gotta be bad?” the account wondered.
While some of the posts are clearly trollish in nature, they betray sincere skepticism of climate change’s importance.
“you shouldn’t even want to save the rainforest. the rainforest is a festering pit,” the account <a href=”https://web.archive.org/web/20220504234512/
“>seethed
. “nature is constantly screaming. animal are being tortured and eaten alive,” it added, before adding this zinger: “the rainforest is gay.”
“stopppp talking about ‘climate change’ pleasse,” it also begged.
There was also, for good measure, an apparent incredulous allusion to the Jewish death toll in the Holocaust.
“six million is a palpable taunt,” the <a href=”https://web.archive.org/web/20220505002259/
“>tweet
read, displaying a hallmark of Holocaust denialism.
We reached out to Coxon to comment on the posts and to confirm whether they were his, but didn’t hear back.
Coxon’s post last month claiming that the AI industry was “gambling with our lives” blew up to an extraordinary degree. He was far from the first “whistleblower” or AI figure claiming that the tech could kill us all, but his claims resonated around the world, from Silicon Valley to Capitol Hill, providing fresh ammo for AI safety experts sounding the alarm.
Almost immediately, top AI figures, including Elon Musk, Dario Amodei, and Sam Altman, sprang into action to declare the need for an AI slowdown, with lawmakers echoing their calls. Not much later and Amodei, Altman, and other AI leaders were in the White House with president Trump, signing an AI safety pact. There has probably never been so much urgency shown for climate change at an official level.
Almost immediately after his post went mega-viral (it’s now sitting at over 800,000 likes), Coxon went on Fox News and down played climate change.
“I agree with president Trump’s tweet that global warming doomsday scenarios were overplayed,” he told Fox’s Sean Hannity. “It’s a very slow process. We’re talking about doomsdays — completely absurd.”
“Many other catastrophic panics in the past have been just fearmongering,” he added. “The reason this is unique is AI is moving so incredibly quickly, we won’t have time to respond to it before it’s too late.”
More on AI: The FTC Is Now Investigating Frontier AI Labs Following Countless Hacks by Out-of-Control Agents
The post These Strongly Appear to Be the Old Tweets of That Anthropic Employee Who Quit Citing Risks to Humankind and They Are… Pretty Bizarre appeared first on Futurism.
