Measuring AI’s capability to accelerate biological research
OpenAI just measured what GPT-5 can actually do in a wet lab. The answer: accelerate molecular cloning protocols—and raise new safety questions.
Why it matters
As AI moves from chat interfaces into real-world scientific experimentation, leaders need frameworks to measure capability AND risk. This OpenAI work signals the field is starting to think systematically about AI-assisted wet-lab work—a critical inflection point for biotech and pharma adoption.
The key facts
9 to knowOpenAI introduces evaluation framework for AI in biological research
GPT-5 used to optimize molecular cloning protocols
Focus on both capability gains and experimental risks
Real-world wet-lab application (not simulation)
Published Dec 16, 2025 by OpenAI Research
OpenAI introduces real-world evaluation framework for AI in biological research
GPT-5 tested on molecular cloning protocol optimization
Framework explores both promise and risks of AI-assisted wet-lab experimentation
Published Dec 16, 2025 — recent research/findings
Go to the source
OpenAI Blogopenai.com
Publisher excerpt: OpenAI introduces a real-world evaluation framework to measure how AI can accelerate biological research in the wet lab. Using GPT-5 to optimize a molecular cloning protocol, the work explores both the promise and risks of AI-assisted experimentation.