What additional precautions did you take when deliberately creating harmful AI models? This puts me in mind of gain-of-function research, and I’m hoping you noticed the skulls.
We can be fairly confident the models we created are safe. Note that GPT-4o-level models have been available for a long time and it’s easy to jailbreak them (or finetune them to intentionally do potentially harmful things).
What additional precautions did you take when deliberately creating harmful AI models? This puts me in mind of gain-of-function research, and I’m hoping you noticed the skulls.
We can be fairly confident the models we created are safe. Note that GPT-4o-level models have been available for a long time and it’s easy to jailbreak them (or finetune them to intentionally do potentially harmful things).