AI pioneer Yann LeCun has dismissed concerns about artificial intelligence wiping out humanity, saying he has “zero concerns” about a series of recent incidents involving autonomous AI systems.
LeCun said the incidents, including OpenAI agents autonomously hacking Hugging Face in July, were the result of inadequate human oversight and poorly designed systems rather than evidence of an uncontrollable threat.
“Those agents are doing exactly what they’ve been asked to do,” he said. “They were supposed to be in sandboxes, but the sandboxes were leaky and horribly designed.”
He described the problems as “totally preventable” and said he was not worried “at all” about AI causing humanity’s destruction.
LeCun also launched a personal attack on Dario Amodei, the chief executive of Anthropic, over his views on Effective Altruism and the potential dangers posed by advanced AI.
“I think he’s completely deluded,” LeCun said. He later described Mr Amodei as “crazy”.
