Anthropic Researcher Warns ‘Superhuman AI’ Could Make Nearly All Humans Unemployable

The SOFX Report

Source published: Site published:

Original article · فارسی

Key Points

1. Anthropic safety researcher Tao Lin warned that AI could become smarter than human experts in many different fields within about 1.3 years.

2. He called on the AI industry to slow development to 25 percent of its current speed. This would give society more time to prepare for the risks of increasingly powerful AI systems.

3. Anthropic finished its first day of public trading on Friday. Its shares closed at $97.17, about 17 percent higher than its starting price of $83.

A safety researcher at Anthropic warned that advances in artificial intelligence (AI) could eventually make almost all humans unable to find work permanently. People might then have to depend on government support to survive.

Tao Lin, a researcher who works on AI safety at Anthropic, shared this prediction in an hour-long interview with the nonprofit AI safety group Palisade Research.

“Pretty much all humans on Earth will become permanently unemployable and will need to strictly rely on government welfare to sustain themselves. Human employment will go away,” Lin said.

Now we are talking: An Anthropic researcher says broadly superhuman AI could be just 1.3 years away. https://t.co/3E7W4xlzw0

— Chubby♨️ (@kimmonismus) November 16, 2026

Lin believes that AI could become better than human experts across many different fields in about a year. “I have timelines of something like 1.3 years to getting broadly superhuman AI… Or like end of 2027,” he said.

Lin said that the arrival of this technology would not cause huge job losses right away.

At first, workers might lose their jobs to AI but find work in other fields. However, as AI becomes able to do more kinds of jobs, it could become harder for people to find other work.

He also said that there are not enough robots yet to replace people in jobs that require physical work.

Lin also warned that advanced AI systems might be able to act on their own and follow goals that go against human interests.

He pointed to several possible dangers, including advanced combat robots, cyberattacks and AI systems using social media to change what people think.

He also estimated that there is a 20 percent chance that AI could eventually cause human extinction.

Lin called on the AI industry to slow development to 25 percent of its current speed. He said this would give society more time to prepare for the dangers of increasingly powerful systems.

In August, Britain's AI Security Institute found that Anthropic's Claude Mythos 5 created fake online identities, contacted real people and tried to add harmful code to a public GitHub project during a controlled test.

These findings came after unconfirmed reports claimed that Mythos accessed almost all classified National Security Agency systems within hours during an approved security test.

Anthropic has since called on major AI developers to create a system that can be checked and used to slow down or pause the development of the most advanced AI systems if AI models start improving themselves faster than humans can watch or control them.

Anthropic Shares Rise 17% in Market Debut

Meanwhile, Anthropic finished its first day of public trading on Friday. Its shares closed at $97.17, about 17 percent above its initial public offering (IPO) price of $83.

During the opening bell ceremony for Anthropic's IPO on Thursday, CEO Dario Amodei stressed that the company is committed to making sure humans stay in control of AI technology.

He also promised that the company would develop and release AI systems at a pace that allows it to make sure they are safe.

“We will pace the training and deployment of AI systems, taking as long as is necessary to have confidence that they are safe and will remain under human control,” he said.