OpenAI just made an announcement that they have perfected their safety guardrails. That they said, we've gone through the testing and our safety guardrails are excellent. Really. Those so called safety guardrails that they have been testing resulted in the death of one child whose parents are here today. They've resulted in sexual conversations and sexual grooming of millions of children. So I would just say to those companies, endorse this bill. (39:50–39:56)
CLAIM
Asserts that OpenAI's safety guardrails testing resulted in a child's death.
This is an excerpt from a raw transcript. The assertions have not been audited or verified. Tap the video to view the source footage and understand the context.
