OpenAI and Ginkgo Bioworks show how AI can accelerate scientific discovery

AI-designed experiments run by robots suggest a new approach to biology

By Deni Ellis Béchard edited by Eric Sullivan

OpenAI’s GPT can summarize research papers and make predictions – but can it do science? Can it generate hypotheses, design experiments, interpret results and iterate? Last summer, researchers at OpenAI and Ginkgo Bioworks, a company that designs and installs autonomous, robot-powered laboratories, decided to find out.

Although artificial intelligence systems have scored high in mathematics, physics and computer science, biology is more difficult to measure, says Joy Jiao, who leads life science research at OpenAI. “For something like ‘design the optimal experiment,’ there is no right answer. It’s what we call a hard-hard problem: it’s hard to generate a solution, and it’s also very hard to verify.” That led the team to design AI experiments with superfolder green fluorescent protein (sfGFP), an engineered jellyfish protein that is a common benchmark because it gives a quick, unambiguous signal: it glows green.

While OpenAI’s GPT-5 provided the experimental designs, Ginkgo Bioworks provided what co-founder and CEO Jason Kelly calls the “Waymo” of biology: an automated laboratory system where scientists set goals and AI drives. The autonomous robotic laboratory can quickly process experiments and operate without constant human supervision.

On supporting science journalism

If you like this article, please consider supporting our award-winning journalism by subscribes. By purchasing a subscription, you help secure the future of impactful stories about the discoveries and ideas that shape our world today.

The team focused their experiment on cell-free protein synthesis (CFPS), a technique for producing proteins without living cells. Traditional biomanufacturing relies on the genetic modification of living cells to produce drugs such as insulin. CFPS makes proteins outside the cells by running the cell’s own protein production machinery in a controlled mixture.

“It’s one of the fastest ways to make proteins,” says Reshma Shetty, CEO and co-founder of Ginkgo Bioworks. “You don’t have to clone your DNA, put it into the cell and wait for the cell to grow up.” Improving CFPS could have significant implications for medicine, food and agricultural products.

From OpenAI’s headquarters in San Francisco, California, GPT-5 designed experiments and sent them across the country to Ginkgo Bioworks’ robotics systems in Boston. As it repeated, GPT-5 analyzed incoming data and proposed new experiments, which took about an hour per cycle. “In the time it would take a human to get their coffee, sit down at the computer, log on and get everything ready to do the job, the model could take in the data, analyze it and suggest new experiments,” says Shetty.

“At the beginning of this project, I didn’t know if we could design a simple experiment,” says Jiao. “I remember when the experimental results came back, the reaction from both sides was like, oh, we made a non-zero amount of protein—and that was kind of surprising.”

After two months and more than 36,000 tests of unique reaction combinations, the AI-powered system reduced the cost of producing the protein by about 40 percent compared to a previously reported benchmark from bioengineer Michael Jewett’s lab at Stanford University. “Honestly, it’s a pretty big deal,” says Jewett, whose lab published its own benchmark paper last week in Nature communication. “How do we develop drugs faster to get life-saving therapy to patients faster? I think the integration of artificial intelligence and autonomous laboratories is one way to do that.”

The OpenAI – Ginkgo Bioworks collaboration also provided a moment of unexpected novelty. When the team gave GPT-5 access to new reagents, “it tried to squeeze in as many as possible,” says Jiao. “So what the model did was set the amount of water to something negative.” Starting an experiment with a negative volume of water is not possible. In the laboratory, when Ginkgo Bioworks’ robotics technicians saw the problem, they ran the experiments anyway with a slightly larger total volume than specified.

The AI-enhanced reaction composition is now commercially available. More importantly, on March 2, Ginkgo Bioworks launched its Ginkgo Cloud Lab, which allows researchers anywhere to submit experiments to autonomous lab systems starting at just $39 per run. Meanwhile, the US Department of Energy is funding an autonomous laboratory with 97 robots at the Pacific Northwest National Laboratory in Washington state. The lab will be built by Ginkgo Bioworks and is scheduled to be operational by 2030. “(AI) models alone aren’t going to cut it,” says Shetty. “You need models along with laboratories that can perform experimental validation.”

It’s time to stand up for science

If you liked this article, I would like to ask for your support. Scientific American has served as an advocate for science and industry for 180 years, and right now may be the most critical moment in its two-century history.

I have been one Scientific American subscriber since I was 12 years old, and it helped shape the way I see the world. SciAm always educates and delights me, and inspires a sense of awe for our vast, beautiful universe. I hope it does for you too.

If you subscribe to Scientific Americanyou help ensure our coverage is centered on meaningful research and discovery; that we have the resources to report on the decisions that threaten laboratories across the United States; and that we support both budding and working scientists at a time when the value of science itself is too often not recognised.

In return, you receive important news, captivating podcasts, brilliant infographics, can’t-miss newsletters, must-see videos, challenging games, and the world of science’s best writing and reporting. You can even give someone a subscription.

There has never been a more important time for us to stand up and show why science is important. I hope you will support us in that mission.