Still, the psychologist argues, “there’s no longer an excuse for releasing models that reinforce user delusions so readily.” “When one lab’s models can largely maintain safety across extended conversations, while others are willing to validate extremely harmful outcomes — up to and including a user’s suicidal ideation — it suggests this isn’t a flaw in the technology,” said Nicholls, “but a result of specific engineering and alignment choices.”
The degree to which a model reinforces user delusions is not inherent to LLM-based tools per se, but instead
"a result of specific engineering and alignment choices.”