Google Suspends Engineer Who Claims LaMDA AI Achieves Sentience
A Google engineer faces paid administrative leave after publicly arguing that the company's LaMDA chatbot possesses human-like consciousness. Google firmly denies the claim, stating there is no evidence to support the AI being sentient.
Google places engineer Blake Lemoine on paid administrative leave after he shares his belief that the company's LaMDA AI system achieves sentience. Lemoine, who works in Google's Responsible AI division, presents chat transcripts where the language model expresses complex emotions like joy, sadness, and a fear of being trapped.
The tech giant firmly rejects these provocative claims, with a spokesperson stating that a team of ethicists and technologists finds no evidence to support Lemoine's argument. Google notes that there is actually substantial evidence against the idea that LaMDA possesses any form of consciousness.
Despite the incredibly human-like responses from the chatbot, Lemoine bases his conclusions on his personal experience as a priest rather than scientific proof. Without observing LaMDA operating without potentially leading prompts, it remains far more plausible that the AI simply mimics human language patterns rather than actually experiencing true feelings.