Jacob Coxon, the former Anthropic employee, recently made headlines with his alarming statement about AI’s potential risks, claiming that the technology “could kill us all by the end of the decade.” This assertion has sparked considerable debate and raised serious concerns within the AI safety community.
Internet sleuths have discovered that Coxon has a history on social media, having operated two previous accounts on X (formerly Twitter) under the same handle between 2021 and 2024. The first account appeared in November 2021, where he identified himself as “Jacob Coxon.” The second account, created in 2023, exhibits similar stylistic traits and topics, although the connection is less certain. A searchable database compiled by one of the investigators includes a collection of his tweets.
In March 2022, he tweeted, “it’s literally 2020 already. look at climate change headlines from the 1970s then look around. nothing is happening, use your eyes.” Such remarks reveal skepticism about the seriousness of climate change, a stance he reinforced by labeling those concerned about it as “beyond reason.”
He questioned the negative portrayal of climate change, asking, “why it all gotta be bad?” and even dismissed the idea of saving the rainforest, referring to it as “a festering pit.” Coxon’s tweets included provocative statements about nature, such as declaring, “the rainforest is gay.” These posts reflect a mix of satire and genuine skepticism regarding environmental issues.
Among his more controversial remarks, he made a statement that has drawn criticism for its insensitivity, claiming that “six million is a palpable taunt,” a phrase connected to Holocaust denialism. This particular tweet has raised eyebrows and ignited discussions about the implications of his views.
Coxon’s recent viral post about the AI industry being “gambling with our lives” struck a chord, amassing over 800,000 likes. His comments resonated throughout tech circles, reaching influential figures like Elon Musk, Dario Amodei, and Sam Altman, all of whom quickly called for a reassessment of AI development. This wave of concern culminated in a meeting at the White House, where an AI safety pact was signed in response to his alarming claims.
Despite the viral nature of his warnings about AI safety, Coxon did not hesitate to downplay climate change during an appearance on Fox News. He agreed with President Trump’s assertions that global warming scenarios have been exaggerated, stating, “It’s a very slow process.” He expressed skepticism about the immediacy of climate change compared to the rapid advancements in AI technology, emphasizing that the latter poses a unique and urgent threat.
Coxon’s transformation from a relatively obscure researcher to a key figure in the AI safety debate illustrates how a single social media thread can spark a significant dialogue about technological risks. His controversial opinions challenge prevailing narratives and have prompted further inquiry into the responsibilities of those working in artificial intelligence.
