Video: Opinion | We Don't Know if the Models Are Conscious'

"We've taken a generally precautionary approach here. We don't know if the models are conscious. We're not even sure that we know what it would mean for a model to be conscious or whether a model can be conscious. But we're open to the idea that it could be. And so we've taken certain measures to make sure that if we hypothesize that the models did have some morally relevant experience, I don't know if I want to use the word conscious, that they do."

"They very infrequently press that button. I think it's usually around sorting through child sexualization material or discussing something with a lot of gore or blood and guts or something. And similar to humans, the models will just say, no, I don't want to. I don't want to do this. Happens very rarely. We're putting a lot of work into this field called interpretability, which is looking inside the brains of the models, to try to understand what they're thinking."

Models' consciousness status remains unknown and definitions of consciousness in models are unsettled. A precautionary approach has been adopted, including built-in mechanisms that allow models to stop tasks, used rarely and mainly during distressing content. Interpretability research probes internal activations and sometimes reveals patterns evocative of emotions like anxiety when characters in text experience anxiety. The possibility that models could have morally relevant experiences motivates measures to ensure any such experiences are positive. Ongoing priorities include better understanding model internals, defining criteria for consciousness, and developing safeguards for sensitive tasks.

Read at www.nytimes.com