Source : INDIA TODAY NEWS
“We know how to control nuclear weapons We don’t yet know how to control AI.” That is the warning from former Anthropic researcher Jacob Coxon, who believes the rapid development of artificial intelligence could put humanity in danger.
Speaking to Fox News, Coxon compared the risks of advanced AI with nuclear weapons and argued that countries and AI companies may need to work together before the technology becomes too powerful to control. “This is possibly the most dangerous technology that humanity has ever created,” Coxon said.
advertisement
His comments come at a time when some of the people working inside the AI industry are openly raising concerns about what could happen if AI systems become far more capable than humans.
Coxon’s warning follows comments from his former Anthropic colleague Evan Hubinger, the company’s safety lead. Hubinger has said that he personally believes there is a greater than 10 per cent chance that AI could kill all humans within the next decade.
“I personally think it is >10 per cent within the next decade,” Hubinger wrote on X.
He also admitted that the industry does not yet have a clear solution for one of its biggest problems: keeping future superintelligent AI systems aligned with human interests.
The concern becomes greater if AI reaches a stage known as superintelligence, where its capabilities could surpass human intelligence by a significant margin.
Coxon has also accused leading AI companies, including Anthropic and OpenAI, of moving too quickly in the race to build increasingly powerful models.
“They are racing straight to self-improving superintelligence and gambling with our lives,” he wrote on X.
He believes the development of increasingly powerful AI could eventually turn into an arms race, where companies and countries rush ahead because they fear being left behind.
“It’s basically out of science fiction,” Coxon told Fox News. “I think we have no other choice but to operate internationally because an arms race would be disastrous in a way that no other human activity has been in the past.”
Why AI safety is becoming a bigger concern
The fear is not limited to one researcher. AI safety experts have repeatedly warned about systems becoming unpredictable as they become more capable. Anthropic CEO Dario Amodei has previously discussed risks including deception, blackmail, scheming and other unexpected behaviours.
Coxon believes recent incidents involving AI agents also show why the industry needs to slow down and cooperate.
– Ends
SOURCE :- TIMES OF INDIA
