OpenAI acquires Orchestra Bio to put AI agents in charge of drug programme operations
OpenAI has acquired Orchestra Bio, a platform that helps biopharma teams coordinate the complex process of drug development, and is bringing Orchestra’s team and technology together with Rosalind, its AI models and tools for life sciences research. Financial terms of the acquisition were not disclosed. OpenAI also gave no timeline for when Orchestra’s agents will show up in its products.
“AI can help take on these specialised tasks and complex workflows,” OpenAI stated in its announcement, adding that the industry is now at an inflection point. The company said developing a new medicine takes roughly 10 to 15 years on average from target discovery to regulatory approval in the United States, and that beyond the science, teams must coordinate programmes across internal departments and external partners, often relying on disconnected systems.
Orchestra Bio was founded in San Francisco by CEO Ashoka Rajendra and CTO Andy Xu in 2022. The company raised about $20 million in total, with a seed round that included Andreessen Horowitz, Definition Capital, Insight Partners and 8VC, according to Caplight. At its August 2025 launch, Rajendra pitched Orchestra to biotech founders as software that came with guidance on financial management and long-term planning.
“Agents can handle planning, orchestration, and analysis,” Orchestra Bio said in a note on its website announcing the deal, adding that scientists and programme leaders would then focus on the decisions that require their expertise. Rajendra and the startup’s team are joining OpenAI.
Discovery was only half the problem
On Orchestra’s platform, scientists and programme leaders delegate the monitoring and tracking of studies and campaigns to agents. OpenAI offered a manufacturing example: a failed batch can delay the start of a clinical trial, and Orchestra’s agents help teams identify which activities and timelines are affected, gather the relevant information and compare ways to respond. TechCrunch’s coverage of the deal also credits the agents with keeping track of finances and resolving clinical trial scheduling problems.
The acquisition fills a gap in the Rosalind stack. Rosalind dates to April, when GPT-Rosalind launched as a reasoning model for biology and drug discovery research, and the platform connects researchers to more than 50 scientific tools and data sources. OpenAI had already secured collaborations with Amgen, Moderna, the Allen Institute and Thermo Fisher Scientific at launch, and its partnership with Novo Nordisk aims to translate the drugmaker’s scale and scientific expertise into faster R&D and commercial execution. Rosalind is positioned to help scientists decide what to test, while Orchestra’s agents handle the downstream operational follow-up needed to turn findings into medicines.
The measurable savings sit in operations
The economics explain the bet. The World Economic Forum estimates that developing a single drug now costs over $2.5 billion, with clinical trials consuming almost half of that, and that nearly 90% of clinical trials fail, often due to poor design, inadequate recruitment and fragmented data systems.
Early evidence on agents in this layer is emerging. A Tufts Centre for the Study of Drug Development analysis found AI agents can accelerate clinical development of cancer drugs by approximately 10 weeks and cut direct operating costs by as much as $5.6 million in late-stage trials, with an experimental treatment across 50 indications seeing net benefits of up to $565 million. Ken Getz, the centre’s executive director, said the agent reduced on-site visits, accelerated enrolment and sped up data lock.
“We have moved past the era of speculation,” said Lisa Moneymaker, chief strategy officer at Medidata, on the company’s 2026 State of AI in Clinical Trials report. The survey of 200 senior decision-makers found 72.9% of respondents with more than 18 months of AI experience report shorter clinical trial timelines, while 92% plan to increase AI spend.
AI labs are now competing for pharma’s workflow
OpenAI is not alone in buying into life sciences. Anthropic acquired AI biotech startup Coefficient Bio for roughly $400 million in April, The Information reported, and launched Claude Science in June, a research platform connecting scientific databases, analysis tools and specialised agents. Amazon launched Bio Discovery, an AI agent that generates and evaluates potential drug molecules, in April.
For pharmaceutical companies, the shift changes what they are buying. Instead of selling models to biopharma teams and relying on them to wire the technology into their workflows, OpenAI can now offer the workflow itself. OpenAI said bringing Orchestra together with Rosalind is a step towards AI supporting scientists across the full process, from exploring ideas to coordinating the work needed to advance treatments.
“Meaningful, sustained disruption to drug development” by AI remains unproven, researchers writing in Nature noted, as Axios reported in April. No fully AI-discovered or AI-designed drug has yet made it through phase 3 trials. Some of the biggest trial challenges also sit outside information processing, including finding the right patients, securing consent and manufacturing and distributing the drug, and human verification of an agent’s work could offset some of the time savings.
Adoption barriers remain high. Medidata’s respondents cited integration complexity (79.5%), model accuracy (77.5%) and weak data foundations (75%) as top obstacles, and 63% mandate human oversight. OpenAI has not said whether Orchestra will remain available as a standalone product or how the acquisition will affect existing customers.
