OpenAI partnered with US-based Los Alamos National Laboratory to study how biological capabilities associated with AI could be used by non-experts to create threats.

Los Alamos National Laboratory is well known for developing the first atomic bomb during World War II.

It stated the partnership with OpenAI would be the “first of its kind” to study AI biosecurity and the ways it can be used in a lab setting.

OpenAI stated the partnership includes biological safety evaluations to study how GPT-4o’s yet to be released voice assistant technology could be used by scientists to support research in bioscience.

While OpenAI highlights the benefits of AI in laboratories, Los Alamos National Laboratory stated previous research “found that ChatGPT-4 provided a mild uplift in providing information that could lead to the creation of biological threats”.

“AI-enabled biological threats could pose a significant risk, but existing work has not assessed how multimodal, frontier models could lower the barrier of entry for non-experts to create a biological threat”, Los Alamos National Laboratory stated.

Erick LeBrun, research scientist at Los Alamos National Laboratory, stated “understanding any potential dangers or misuse of advanced AI related to biological threats remain largely unexplored”.

He noted the work with OpenAI will establish “a framework for evaluating current and future models”.