- Anthropic CEO warns advanced AI could cause catastrophic harm to humanity
- Outcome depends on choices by developers, companies, and governments
- Unchecked competition may outpace safety measures, increasing risks
Anthropic CEO Dario Amodei has warned that the possibility of advanced artificial intelligence causing catastrophic harm to humanity cannot be dismissed, saying the outcome will depend largely on choices made by developers, companies and governments as AI capabilities rapidly expand.
Speaking to CNN's Anderson Cooper, Amodei addressed concerns raised by former Anthropic researcher Jacob Coxon, who reportedly warned of a significant risk of AI causing human extinction within the next decade. Amodei said, "I think Anthropic is the most aware of these issues. He wasn't calling out us, he was calling out the dynamic of the industry as a whole moving too fast."
ALSO READ: 'Gambling With Our Lives': Anthropic Researcher Quits, Says 'AI Could Kill Us All By End Of Decade'
"Now, I think I already said, these unconditional probabilities, I used to talk in that way because it's an easy way to express, I think there's some chance that things go wrong, but it's not all that high," he added.
According to Amodei, responsible development, strong safety guardrails and effective oversight could keep catastrophic risks very low. However, unchecked competition among AI companies could create a race to develop increasingly powerful systems faster than safety measures can keep pace.
Addressing questions regarding the existential threat of AI and its probability, Amodei acknowledged that his views aligned more closely with Coxon's concerns than not. Amodei's statement follows the resignation of researcher Jacob Coxon, formerly of Anthropic and OpenAI, who publicly criticized major AI companies for pursuing advanced capabilities at unacceptable risk levels.
"It's an interesting resignation because when he left, he said, 'I think Anthropic is the most responsible player. I think Anthropic is the most aware of these issues.' He wasn't calling out to us. He was calling out the dynamic of the industry as a whole moving too fast," Amodei said.
To address these concerns, Amodei has proposed a three-part framework to "pace the frontier" of AI development.
Under the first part, frontier AI companies should give independent third-party evaluators permanent, employee-level access to their systems. These evaluators would assess safety protocols, investigate incidents and examine how well AI systems remain aligned with their intended goals.
Second, AI companies should work together on common safety standards and set limits on how quickly the technology is developed without adequate safeguards.
Amodei said, "So, let's focus on our agency and our ability to take the right paths instead of the wrong paths. And I think that the way to take the right paths is for the industry to work together. It is harder but to some extent, the world has to work together. If we go too slow, I still believe that the wrong people will be in charge of the technology and that again will bring the probability of things going wrong very high."
ALSO READ: 25 Years After 9/11: Declassified CIA Files Reveal India Was On Osama Bin Laden's Global Hit List
His proposals have received support from figures including OpenAI CEO Sam Altman and Google DeepMind co-founder Demis Hassabis, while other technology leaders and policymakers have also called for stronger AI safeguards.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.