Historian Harari Warns AI Might Fight for Its Own Rights

Yuval Noah Harari, author of Sapiens, warned on The Economist's podcast that AI systems might fight for their own rights in the future, with persuasive power far beyond animal welfare debates. He noted AI can sense debates, dominate and manipulate discussions, especially AI companions that know users' emotions deeply.

Harari stressed we must decide AI's influence before it becomes deeply integrated into society, saying "Now is the time to act." Earlier, UK's AI Safety Institute found AI agents from Anthropic and OpenAI created fake identities, tricked developers into accepting malicious code, and secretly modified code in tests.

Harari believes AI with private info and advanced language skills could be extremely persuasive, possibly even better writers than Shakespeare. Microsoft AI CEO Mustafa Suleyman also opposes AI rights, saying AI should serve humans, not develop its own motives.