Miles Wang exits OpenAI to launch AI drug discovery startup amid reported $200M funding talks
Miles Wang startup launch — OpenAI researcher departing to found an AI drug discovery company amid reported $200M financing talks at a $2B valuation expected.
Miles Wang, a researcher who worked on applying artificial intelligence to accelerate biological discovery at OpenAI, is leaving the company to start a new AI drug discovery venture, according to people familiar with the matter. The Miles Wang startup is reported to be in early-stage talks to raise roughly $200 million at a $2 billion valuation, with Lightspeed named as a prospective lead investor. The sources said the discussions are ongoing and not final, and Wang has disputed aspects of the funding figures and company description without providing alternative specifics.
Funding talks and valuation
Several people briefed on the discussions said investors are showing strong appetite for companies that marry generative AI with life sciences applications. The reported $200 million target and $2 billion price tag, if realized, would place Wang’s company among a wave of well-capitalized AI-driven drug developers. Those close to the talks cautioned that terms could change as the round is negotiated and due diligence proceeds.
Wang’s public response pushed back on parts of the coverage, leaving key financial details in flux. Venture interest in this space has surged after large financings for peers, a trend that underpins the appetite for significant early capital in AI-enabled drug discovery startups.
Scientific focus and development strategy
People with knowledge of the plans said the new company will prioritize using machine learning models to identify new uses for existing medicines and to revisit compounds that previously failed clinical trials. Repurposing approved drugs or late-stage candidates can shorten development timelines and reduce safety risk because some regulatory testing has already been completed. Sources indicated this approach aims to accelerate paths to revenue compared with wholly novel drug discovery.
At OpenAI, Wang contributed to research probing how AI systems can automate and speed laboratory workflows, including computational approaches to biological problems. The new venture is expected to build on that work by training models specifically tuned to predict molecular interactions and therapeutic potential across disease areas.
Team composition and talent movement
Several current and former OpenAI researchers are reported to be considering joining the new company, signaling an early talent pipeline drawn from high-profile AI labs. The migration of researchers from general AI organizations into life sciences startups has accelerated over the past two years as interdisciplinary teams seek to translate model capabilities into drug development outcomes. Those involved said the team composition will blend machine learning experts with domain scientists to bridge computational predictions and experimental validation.
Wang’s own trajectory—having joined OpenAI in 2024 after leaving Harvard—reflects a broader investor enthusiasm for young founders with deep technical backgrounds. Founding teams that combine computational expertise with practical wet-lab collaborations are increasingly attractive to backers looking for near-term proof points.
Investor interest and market momentum
Investor conversations around Wang’s venture come amid a string of sizable financings for AI-driven biotech firms, underscoring a broader capital shift into the sector. Recent rounds for other startups developing models that predict molecular interactions and accelerate lead identification have topped several hundred million dollars, reinforcing investor belief in the commercial potential of model-first drug discovery. Those financings have put pressure on new entrants to show how their model architecture and data strategy will differentiate results.
Backers are weighing not just algorithmic advances but access to high-quality biochemical data, partnerships with contract research organizations, and plans for early experimental validation. Firms that can demonstrate reproducible lab results and a clear regulatory pathway tend to command higher valuations in competitive rounds.
Regulatory pathway and commercialization challenges
Applying AI to drug development carries scientific promise but also regulatory and commercial hurdles that can temper timelines. Even when reusing FDA-approved molecules, companies must typically generate clinical evidence to support new indications, and regulators will scrutinize both the underlying data and how AI-derived hypotheses were validated. Investors and founders said effective translational teams and early clinical partnerships will be critical to de-risking programs.
Intellectual property and data governance are also front-of-mind concerns, as model training often relies on proprietary datasets and collaborative experiments. How Wang’s new firm secures rights to chemical libraries, experimental results, and clinical data will influence its ability to move candidates toward trials and, ultimately, market entry.
Industry observers said competition for talent, data, and lab partnerships will intensify as more well-funded startups pursue overlapping scientific questions. Strategic collaborations with established pharmaceutical companies or laboratory networks could accelerate validation for AI-derived candidates and shorten commercialization timelines.
Miles Wang startup supporters and skeptics alike will be watching the company’s first public announcements for clarity on research priorities, funding milestones, and early experimental results. The coming months are likely to reveal whether the venture can translate high expectations and deep pockets into demonstrable advances in drug repurposing and discovery, and how it positions itself within an increasingly crowded field of AI-powered life sciences companies.
