Altman and Amodei Walk Back AI Job Apocalypse Predictions Ahead of IPOs

OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei — two of the most vocal prophets of an AI-driven white-collar job apocalypse — are now walking back their predictions, just as their companies prepare for blockbuster IPOs. In separate recent interviews, both admitted their earlier warnings were overblown, echoing a stance long held by Goldman Sachs CEO David Solomon.
Altman: 'I was pretty wrong'
Altman, in a conversation with Commonwealth Bank of Australia CEO Matt Comyn, said he was "pretty wrong" about AI's economic impact — a reversal from his June 2025 warnings that entry-level roles were at serious risk. He even tried delegating his Slack and email responses to AI, but eventually reverted to handling them himself. "We really do care about our interactions with people," Altman said. "This thing … is not something that I can imagine myself outsourcing to an AI anytime soon. It really updated me to thinking that the jobs picture is likely to be very different than we thought."
Amodei: Automation multiplies, not destroys
Amodei, who once claimed AI could eliminate 50% of white-collar jobs, now frames automation as a multiplier of output: "If you automate 90% of the job, then everyone does the 10% of the job … And the 10% kind of expands to be 100% of what people do and kind of 10-times their productivity."
Solomon: Vindicated by history
Solomon, who consistently argued the panic was overblown, points to U.S. employment data: 145% growth since 1962, with data center construction alone adding 200,000 jobs since 2022. He cites a 2018 study by Nobel laureate Daron Acemoglu that backs his claim that AI's displacement effects may be limited.
Both OpenAI and Anthropic are reportedly preparing to launch their respective IPOs this year, each with an estimated valuation of $1 trillion.
📖 Read the full source: HN AI Agents
👀 See Also

Top AI Models Show Performance Gap in Non-English Languages
A recent analysis shows leading AI models perform worse in languages other than English, with the article receiving 16 points and 3 comments on Hacker News.

Stanford CS 25 Transformers Course Opens to Public with Live Streaming
Stanford's CS 25 Transformers seminar is now open to the public with lectures starting January 23, 2025, at 4:30-5:50pm PDT, available in-person at Skilling Auditorium or via Zoom, with recordings posted online.

Claude Code Randomly Becomes Risk-Averse, Demanding Permission on Routine Tasks
A user reports that Claude Code intermittently shifts from autonomous execution to requiring excessive permissions, even on daily, unchanged workflows like rebuilding a monorepo and running tests.

MiMo-V2.5-Pro Benchmarked: Strong Social Deduction Reasoning, Good Value vs K2.6
MiMo-V2.5-Pro competes with Kimi K2.6 in autonomous Blood on the Clocktower games, with a lopsided 88% Good / 48% Evil win rate, costs $0.99/game at 183k output tokens, and is practical with 2-3 hour matches.