Ex-Anthropic researcher warns about the impossibility of complete control over AI
Quick Look
- Former Anthropic researcher Jacob Coxon said artificial intelligence cannot yet be fully controlled, warning of the risk of an apocalyptic scenario by the end of 2027 due to the race among tech corporations.
- US Treasury Secretary Scott Bessent believes that if threats are confirmed, the development of neural networks should be slowed down.
AI-generated summary
Why It Matters
Jacob Coxon previously worked at Anthropic, an artificial intelligence security research company, and left due to concerns about the uncontrolled development of AI. His statements come in the context of a growing debate about the regulation of advanced technologies.
Brief retelling from RIA II
Jacob Coxon said that artificial intelligence cannot yet be completely controlled.
Coxon fears that the race among tech corporations could lead to an apocalyptic scenario as early as the end of 2027.
US Treasury Secretary Scott Bessent believes that if fears about the dangers of technology are justified, the development of neural networks should be slowed down.
NEW YORK, October 5 - RIA Novosti. Former Anthropic researcher Jacob Coxon said that artificial intelligence cannot yet be completely controlled.
Coxon previously announced his departure from the AI industry, fearing that the race among technology corporations is bringing the world closer to the “point of no return” and could lead to an apocalyptic scenario as early as the end of 2027.
"We don't yet know how to control any artificial intelligence system. So any AI system that we create, or perhaps more accurately, grow, we do not fully control," he said during a special hearing in the New York City Council.
What to Watch
AI outlook — possibilities, not facts
The development of neural networks may be slowed down if threats from the US authorities are confirmed
Possible · Within months
The debate on AI regulation will intensify by the end of 2027
Likely · Within years
Open Questions
- What specific measures does Coxon propose to reduce the risks of AI?
- How long will it take to develop effective control mechanisms for AI?
- Which tech corporations does Coxon think are most responsible for accelerating the race?
