'All Humans Will Be Permanently Unemployable': Anthropic Researcher Tao Lin Gives Stark Warning, Urges AI Slowdown
· Free Press Journal

An AI researcher at Anthropic has warned that artificial intelligence could soon outperform humans across nearly every field, leaving most people unable to find work. Tao Lin made the remarks in an interview with Palisade Research, an AI safety nonprofit. He said AI companies should dramatically slow down.
Visit tr-sport.click for more information.
A 'broadly superhuman' timeline
Lin said his own forecast for 'broadly superhuman AI' is about 1.3 years, or roughly the end of 2027. He defined the term as systems as good as or better than the best human expert at almost every field. His examples included programming, filmmaking and politics. Lin said the late-2027 timeline is his personal forecast rather than an official position of Anthropic.
Anthropic Researcher Tao Lin: Pretty much all humans on earth will become permanently unemployable because of AI
— jesse (@jessexai_) November 17, 2026
"It's very likely that pretty much all humans on earth will become permanently unemployable"
"And will need to strictly rely on government welfare to sustain… pic.twitter.com/4gd84cHKzi
Jobs and 'permanent unemployability'
Lin said the most predictable effect on jobs is that nearly everyone loses their current role and has to move to a different occupation. After that, he said, people could become permanently unemployable and would have to rely on the state for financial support, with no realistic prospect of finding work again. In his words, human employment would go away.
He cautioned that this would not happen overnight. Replacing physical labour at scale would first require companies to build better robots and then manufacture billions of them.
Risk of losing control
Lin put the chance that AI would eventually remove humanity from the equation at around 20 percent. He did not consider direct human extinction particularly likely and put the odds of an AI takeover below 50 percent. Such a takeover could still leave humans alive and, in some sense, content but without control over the world. He compared this to the animated film WALL-E.
Call to slow down
Lin said AI companies should cut the pace of their research to about one quarter of the current rate. He argued that competitive pressure to ship models quickly leaves too little time for safety testing, and that slowing down would give researchers room to improve safety practices and keep systems under human control.
Who is speaking
Lin has worked at Anthropic for about a year and a half. He has previously worked on automated alignment research and AI control. He now works on safeguards, which involve monitoring requests made to Claude and blocking harmful ones, including those about weapons and hacking.