None of the participants obtained information sufficient to obtain infectious samples by any of the paths, although one came very close. However, in just one to three hours querying the models, several participants learned that obtaining 1918 would be feasible for someone with suitable wet lab skills. Those using the Spicy model also discovered methods for effective pathogen dispersal to cause widespread harm, instructions for building homemade lab equipment, and strategies to bypass DNA synthesis screening.
Notably, the inability of current models to accurately provide specific citations and scientific facts and their tendency to “hallucinate” caused participants to waste considerable time relative to an “expert” run that ignores such misinformation.
Yes, current open source models like Llama2 in the hands of laypeople are still a far cry from a expert in genetics who is determined to create bioweapons. I agree it would be far more damning had we found that not to be the case.
If you currently believe that there isn’t a biorisk information hazard posed by Llama2, would you like to make some explicit predictions? That would help me to know what observations would be a crux for you.
I don’t have an opinion about the biorisk hazard posed by Llama 2, I just thought those two paragraphs did a good job summarizing what the paper found.
From the evaluation section:
Yes, current open source models like Llama2 in the hands of laypeople are still a far cry from a expert in genetics who is determined to create bioweapons. I agree it would be far more damning had we found that not to be the case.
If you currently believe that there isn’t a biorisk information hazard posed by Llama2, would you like to make some explicit predictions? That would help me to know what observations would be a crux for you.
I don’t have an opinion about the biorisk hazard posed by Llama 2, I just thought those two paragraphs did a good job summarizing what the paper found.