OpenAI and Ginkgo Bioworks Run 36,000 AI-Designed Biology Experiments
OpenAI's GPT-5 designed and ran 36,000 biology experiments in two months—achieving a 40% cost reduction in producing proteins—demonstrating that AI can do more than summarize papers: it can design, execute, and iterate

OpenAI and Ginkgo Bioworks Run 36,000 AI-Designed Biology Experiments
OpenAI's GPT-5 designed and ran 36,000 biology experiments in two months—achieving a 40% cost reduction in producing proteins—demonstrating that AI can do more than summarize papers: it can design, execute, and iterate real science.
The collaboration between OpenAI and Ginkgo Bioworks, a synthetic biology company with autonomous robot-run labs, shows "AI doing real experimental science: designing experiments, running them, and learning from the results," according to Jason Kelly, Ginkgo's co-founder and CEO.
"AI combined with autonomous labs is needed to keep the United States competitive in science worldwide," Kelly said.
The team focused on cell-free protein synthesis (CFPS), a technique that produces proteins outside living cells. They used superfolder green fluorescent protein (sfGFP) as a benchmark because it provides a fast, unambiguous signal—it glows green when produced successfully.
From OpenAI's San Francisco office, GPT-5 designed experiments and sent them to Ginkgo's robotic systems in Boston. Each experimental cycle—design, execution, analysis, and iteration—took about an hour.
"At the beginning of this project, I didn't know if we could design a single experiment," said Joy Jiao, who leads life sciences research at OpenAI. "I can remember when the experimental results came back... we made a non-zero amount of protein—and that was somewhat surprising."
After two months and over 36,000 tests, the AI-driven system reduced the cost of producing the protein by about 40% compared with a previously reported benchmark from Michael Jewett's lab at Stanford. Jewett called the result "a pretty big deal" and noted that "the integration of artificial intelligence and autonomous labs" could speed up medicine development.
The AI-improved reaction composition is now commercially available. More broadly, the collaboration points to a new model for scientific discovery: AI systems paired with robotic labs that can run experiments without constant human oversight.
Sources
- scientificamerican.com— Scientific American
- prnewswire.com— PR Newswire
Share
Related Articles
Stay in the loop
Get the best frontier systems analysis delivered weekly. No spam, no fluff.
