OpenAI has partnered with the Los Alamos National Laboratory to study the ways artificial intelligence frontier models can assist with scientific research in an active lab environment. Established in 1943, the New Mexico facility is best known as home to the Manhattan Project and the development of the world’s first atomic bomb. It currently focuses on national security challenges under the direction of the Department of Energy. As part of the new partnership, the lab will work with OpenAI to produce what it describes as a first-of-its-kind study on the impact of artificial intelligence and biosecurity.

“The initial experiment involves using AI to help someone who might not be skilled in molecular biology to perform basic biomedical tasks — in this case, helping genetically engineered E. coli bacteria produce insulin,” Axios reports, adding that OpenAI believes the experiment “could help scientists figure out which areas of research GenAI systems could most help.”

“This is a real-world setting where scientists would actually use this model for biological work, and that’s very exciting,” Tejal Patwardhan, a technical lead on OpenAI’s preparedness team told Axios, noting that “evaluating them in a setting where scientists actually work is one of the first steps toward realizing their potential.”

The team’s work will build upon previous efforts “and follow OpenAI’s Preparedness Framework, which outlines an approach to tracking, evaluating, forecasting and protecting against emerging biological risks,” per a news announcement from LANL.

In January, Patwardhan was the lead author on the OpenAI research paper “Building an Early Warning System for LLM-Aided Biological Threat Creation.”

While AI deployment has presented many societal rewards, “rapid advancement in AI’s capabilities have also raised fears among lawmakers and even some tech executives that the technology could be used by bad actors to develop bioweapons,” writes Bloomberg, noting OpenAI’s January study indicates the earlier GPT-4 model “poses ‘at most’ a slight risk of helping people create biological threats, based on early tests.”

Concurrent with the OpenAI news, Los Alamos has established an internal AI Risks and Threat Assessments Group (AIRTAG) “to focus on developing strategies to understand benefits and mitigate risks and help promote the secure deployment of AI tools,” according to the announcement.

Bloomberg says the partnership will also explore how OpenAI’s latest model, GPT-4o, “can be used to support and troubleshoot lab tasks,” including “how GPT-4o’s unreleased voice assistant technology can help scientists with their research.”

The Microsoft-backed startup in recent weeks announced health and biotech deals with Moderna and Color Health.