OpenAI unveils GPT-Rosalind to accelerate life sciences research

Friday 17 April 2026 - 07:50

By: Dakir Madiha

OpenAI unveils GPT-Rosalind to accelerate life sciences research

OpenAI has introduced GPT-Rosalind, a reasoning model designed for life sciences research, marking the company’s first major move into scientific AI tools focused on genomics, drug discovery, and protein engineering. The release came alongside a broad update to its Codex coding agent, which gained more than 90 new plugins announced the same day.

Named after British chemist Rosalind Franklin, whose X-ray crystallography work contributed to the discovery of DNA structure, the model is built to help researchers handle large datasets and fragmented scientific fields. It aims to support scientists in extracting insights from complex biological data and translating research findings into medical applications.

OpenAI said the current drug development process in the United States typically takes 10 to 15 years from identifying a therapeutic target to regulatory approval, with only about one in ten drugs entering clinical trials ultimately approved. The company stated that the model could help accelerate some of the most time-consuming stages of this process.

In benchmark tests measuring understanding of proteins, DNA, and chemical reactions, GPT-Rosalind outperformed earlier OpenAI systems. In one evaluation involving working scientists, it exceeded 95 percent of human experts in predicting the function of certain RNA sequences. OpenAI also released a dedicated life sciences research plugin that connects the model to more than 50 scientific databases, enabling access to protein structures, DNA sequence analysis, and academic papers through a single interface.

Access to GPT-Rosalind will remain restricted due to biosafety concerns. OpenAI said only approved research organizations will be allowed to use the model, with initial partners including Amgen, Moderna, the Allen Institute, and Thermo Fisher Scientific. This approach reflects a broader shift toward controlled access for advanced AI systems, similar to other restricted models developed for cybersecurity research.

The announcement follows recent policy discussions emphasizing the role of artificial intelligence in life sciences. OpenAI has called for improved access to medical data and greater recognition of advanced AI as a strategic research asset. Its foundation also plans to allocate at least $1 billion this year to AI-related initiatives, with a focus on disease research and treatment.

Separately, OpenAI expanded Codex beyond code generation by adding more than 90 plugins, including integrations with CircleCI, GitLab, and Microsoft tools. The update positions Codex as a broader software engineering agent capable of operating across developer workflows.