Avertismentul unui cercetător Anthropic: "Inteligența artificială ar putea ucide omenirea în următorul deceniu. Credem cu toată convingerea asta"

Three researchers from Anthropic warn that uncontrolled superintelligent AI could pose catastrophic risks to humanity within the next decade. Jacob Coxon, who resigned from Anthropic, criticized the industry's rapid development of self-improving AI systems without adequate safety measures. He claims that concerns about potential catastrophic scenarios are shared among high-level executives in AI companies. Evan Hubinger and Samuel Marks, also from Anthropic, echoed these fears, suggesting that the likelihood of AI causing human extinction is significant. Anthropic defended its approach, emphasizing transparency about the risks and its commitment to safety measures in AI development.