Friday, 23 May, 2025

Anthropic CEO Asserts AI Models Hallucinate Less Than Humans, Defends Path to AGI

At Anthropic's inaugural developer event, CEO Dario Amodei claimed that AI models hallucinate less frequently than humans, though in more unexpected ways. He argued that hallucinations don't hinder the pursuit of artificial general intelligence (AGI), which he anticipates could arrive by 2026. While acknowledging challenges, Amodei emphasized the need for improved interpretability to understand AI decision-making processes.