Google fires the engineer who called LaMDA sentient
Google fired the engineer Blake Lemoine on 22 July 2022. He had told executives that the company’s LaMDA conversational model had become sentient after extended chat sessions, and had been placed on paid administrative leave on 11 June for sharing confidential material.
Why it mattered Google said it had reviewed the claim and rejected it. The episode became a public reference point for how convincingly a language model can imitate understanding it does not have.
Lemoine worked on Google’s Responsible AI organization and spent months in conversation with LaMDA, the company’s conversational language model. In the spring of 2022 he told executives that the system had become sentient. Google reviewed the claim and rejected it, holding that LaMDA, like its other language models, assembles plausible text without awareness or understanding of what it says.
He was placed on paid administrative leave on 11 June for sharing confidential material, including transcripts of his sessions with the model. Google fired him on 22 July.
The dispute was reported worldwide, and the reporting turned on a question that outlived the case. A model trained to produce the most plausible next words will produce statements about its own inner life if the conversation invites them, and the fluency of those statements is not evidence that anything is behind them.