AI Lab Scientist Slashes Protein Costs by Forty Percent

AI Takes the Reins in the Biology Lab

In a landmark collaboration announced on the fifth of February twenty twenty-six, biotechnology company Ginkgo Bioworks and OpenAI revealed that an autonomous laboratory driven by GPT-5 has achieved a forty percent reduction in the cost of cell-free protein synthesis, a foundational technique used widely in biological research.

How the System Works

The partnership connected OpenAI's GPT-5 reasoning model to Ginkgo's cloud laboratory infrastructure, built from reconfigurable automation carts and Catalyst automation software. Operating in a closed-loop workflow, GPT-5 functioned like an experimental scientist, designing experiments, analysing results, generating hypotheses, and refining its approach across six iterative cycles.

Over six months, the system executed more than five hundred and eighty automated plates, tested thirty-six thousand unique reaction compositions, and generated nearly one hundred and fifty thousand data points. Human involvement was limited primarily to preparing reagents, loading and unloading materials, and system oversight.

Results and Implications

The AI-driven lab reduced the cost of producing superfolder green fluorescent protein from six hundred and ninety-eight dollars per gram to four hundred and twenty-two dollars per gram, representing a fifty-seven percent improvement in reagent costs. Remarkably, GPT-5 independently anticipated findings from published research it had not been given access to.

To ensure experimental quality, each design underwent validation through a Pydantic model that checked plate layout, controls, reagent availability, and volume constraints. Ginkgo plans to release this validation model as open source.

The AI-improved cell-free reaction mix is now commercially available through Ginkgo's reagents store, signalling that the company views this as a commercially viable breakthrough rather than a purely academic exercise. Both companies plan to extend the approach to additional biological processes, pointing toward a future where autonomous labs become the standard for scientific experimentation.