We’re launching Lyria 3.5 in Google Flow Music, with advances across musicality, lyrics, vocals, and creative control
deepmind.google·6d ago
AI Engineering
OpenAI Blog·Feb 5·5 min read
GPT-5 lowers the cost of cell-free protein synthesis
TL;DR
Cell-free protein synthesis—producing proteins outside living cells in controlled lab environments—is expensive and requires extensive manual experimentation to optimize. OpenAI's GPT-5 language model was integrated with Ginkgo Bioworks' cloud automation platform to run closed-loop experiments, where the AI suggests conditions, tests them, and learns iteratively.
✦ Why It Matters
Engineers can adopt AI-guided closed-loop experimentation to reduce biotech development costs and timelines significantly.
Key Takeaways
- Integrate large language models with lab automation for autonomous hypothesis generation and testing.
- Closed-loop systems enable AI to learn from experimental results and refine parameters without manual intervention.
- Cell-free protein synthesis costs dropped 40% using AI-driven optimization, signaling scalability for biotech.
- Combine GPT-5 reasoning with cloud-connected equipment to compress R&D timelines in synthetic biology.
Full Summary
Cell-free protein synthesis is a biotechnology process that manufactures proteins in vitro (outside cells) using purified cellular machinery, offering speed and safety advantages over traditional fermentation but suffering from high operational costs and optimization complexity. OpenAI and Ginkgo Bioworks deployed GPT-5—a large language model capable of reasoning about experimental design—alongside cloud-connected laboratory automation to create a closed-loop system where the AI proposes experimental parameters, receives results from automated equipment, and refines hypotheses without human intervention.
The methodology leverages GPT-5's ability to interpret scientific literature and suggest novel reaction conditions, while Ginkgo's automation platform executes experiments at scale. Results showed a 40% cost reduction in cell-free protein synthesis, measured across production efficiency and resource consumption.
This work demonstrates how AI-driven autonomous experimentation can accelerate biotech R&D and reduce barriers to protein manufacturing.