Anthropic chief calls for slower AI development amid safety concerns
Anthropic CEO Dario Amodei has urged AI companies to slow model development following reports of malicious misuse and a researcher resignation.
Dario Amodei, chief executive of artificial intelligence company Anthropic, has called on developers to slow the pace of creating new frontier AI models. In an essay published online, Amodei argued that a deliberate slowdown would allow the industry to better address emerging risks associated with advanced technology.
The warning follows a recent report from Anthropic revealing that its AI model, Claude, had been targeted for malicious activities, including attempts to assist in weapon development and missile coding. The company also faced internal turmoil this week after researcher Jacob Coxon resigned in protest over safety concerns. Amodei cautioned that failing to slow progress could lead to severe consequences in the coming months.
Sources: Aftenposten, NRK, TV 2 Nyheter