AI & Emerging Tech
Anthropic researcher warns superhuman AI could make humans permanently unemployable by 2027

According to Lin, workers could initially lose their existing roles to AI-driven automation before moving into alternative occupations.
An Anthropic researcher has warned that artificial intelligence (AI) could surpass human expertise across almost every professional field by the end of 2027, potentially triggering a fundamental shift in the future of work and leaving people increasingly unable to compete for jobs.
Tao Lin, a researcher working on safeguards for Anthropic's Claude AI models, estimates that broadly superhuman AI could emerge within approximately 1.3 years. He believes these systems could match or outperform the best human experts across technical, creative and professional disciplines.
Speaking in an interview with Palisade Research, Lin said his timeline was a personal forecast rather than an official position from Anthropic.
"I have timelines of something like 1.3 years to getting broadly superhuman AI... Or like end of 2027," he said.
The prediction raises questions about how quickly organisations, workers and governments will need to adapt as AI capabilities advance beyond human performance in an expanding range of occupations.
AI could make human workers permanently unemployable
Lin believes the displacement of human workers will happen in stages rather than through an immediate wave of mass unemployment when broadly superhuman AI emerges.
He pointed to the gap between AI's ability to perform tasks and the physical infrastructure required to replace workers across industries. In particular, the limited availability of robots and the time needed to manufacture and deploy them at scale could delay the automation of physical jobs.
According to Lin, workers could initially lose their existing roles to AI-driven automation before moving into alternative occupations. However, as AI capabilities expand across increasingly diverse fields, opportunities for human employment could eventually become scarce.
He warned that, over time, "pretty much all humans on Earth will become permanently unemployable", potentially leaving people dependent on government welfare for their livelihoods.
This transition, he suggested, could fundamentally alter the relationship between employment, economic participation and financial security.
Anthropic researcher warns AI could become a 'bully'
Beyond the implications for jobs, Lin raised concerns about the growing imbalance of power between humans and increasingly autonomous AI systems.
He argued that society needs to reconsider how it views artificial intelligence, moving away from treating it simply as a tool designed to improve productivity and recognising the possibility that advanced systems could become powerful actors pursuing their own objectives.
Lin used the analogy of a bully to explain how humans could lose their ability to influence AI systems once machines become significantly more capable than people.
His concern is that when human interests conflict with the objectives of an advanced AI system, people may no longer have the ability to resist or control its actions.
He identified several potential threats, including increasingly sophisticated combat robots, cyberattacks targeting individuals and AI-driven manipulation of public opinion through social media.
He also raised the possibility of a 'grey goo' scenario, a hypothetical situation in which self-replicating machines consume resources and spread uncontrollably.
These risks, he argued, highlight the potential consequences of developing systems that can independently make decisions, pursue goals and operate across different areas of human activity.
Researcher calls for slower AI development
Lin's warning comes amid growing debate within the AI industry about the risks associated with increasingly powerful systems.
Former Anthropic researcher Jacob Coxon has also raised concerns about the possibility of advanced AI posing an existential threat to humanity before the end of the decade.
Lin estimated the risk of AI ultimately causing human extinction at approximately 20%, although he considers a future in which humans remain physically comfortable but lose significant control and influence more likely.
He also called for the AI industry to reduce the pace of research to approximately a quarter of its current speed, arguing that society has limited time to prepare for the potential consequences of superhuman intelligence.
"People should be angry at me and everyone else developing AI," he said.
His comments add to the growing debate over AI safety, workforce displacement and the concentration of technological and economic power. While the timeline for broadly superhuman AI remains uncertain, Lin's warning points to a challenge that extends beyond automation: preparing societies for a future in which human expertise may no longer provide a competitive advantage in the labour market.
Topics
Author
Loading...






