Introducing Edison Scientific

Two years ago, we launched FutureHouse with the goal of building an AI Scientist. At that time, we expected that it would take 10 years to build agents that were able to do scientific research even semiautonomously, let alone autonomously.

It is often said that people overestimate what they can do in 2 years and underestimate what they can do in 10. These days, with the extraordinary pace of AI research, my impression is that we may actually overestimate what we can do in 6 months, and underestimate what we can do in 2 years. In the past two years, we have made more progress than we had ever thought possible. We released the first AI agent that beat humans at real-world literature search tasks; built one of the first dedicated data analysis agents; demonstrated that we can train agents using reinforcement learning on tasks as diverse as literature search, protein engineering, and chemical reasoning to beat frontier models by up to 50 points; launched one of the first AI Scientist platforms; demonstrated that we could combine literature search, data analysis, hypothesis generation, and experimental planning in a single system to make new, therapeutically important discoveries; and have now demonstrated that our integrated AI Scientist systems can do in a day what would take humans months otherwise. This has all been possible because, with few exceptions, virtually everything we have tried has “just worked” (a very surprising and welcome development coming from wet lab biology). As a result, the vision that we laid out for an AI Scientist in late 2023 feels immediately within reach.

As we have made progress, interest from potential commercial partners has increased in an extraordinary way. Since we launched our platform in May 2025, we have been inundated with requests for rate limit increases that we have not been positioned to support. We have received inbound interest in our agents from VP- or C-level executives at 6 of the top 10 pharma companies, to say nothing of the numerous smaller biotechs that have reached out. Figuring out how to deploy our technology and meet this interest is a critical part of achieving impact. However, building a product, implementing payment systems, going to market, and supporting customers is not the right use of our philanthropic funding, and we can leverage much more capital much more quickly if we use for-profit funding for those purposes, while reserving our non-profit funding for basic research that cannot be funded otherwise. For this reason, it became apparent to us that scaling to meet this demand would require a commercial structure.

Today, therefore, we are launching Edison Scientific, a new commercial spinout that will focus on further developing and deploying our AI Scientist for commercial applications. A portion of the team at FutureHouse is moving to Edison to support this transition; FutureHouse will continue to develop AI Scientists to accelerate foundational research in biology. Andrew and I will run both FutureHouse and Edison for the time being, with strict controls in place to avoid self-dealing or private benefit and manage conflict of interest. Edison is committed to maintaining a generous free tier, in line with FutureHouse’s commitment to accelerating the scientific community, while also ensuring that power users who need higher rate limits or additional features can pay for those features.

The launch of Edison Scientific is a major milestone for FutureHouse as an organization, and is an extremely exciting moment for our pursuit of the dream of automated scientific research. We are extremely grateful to all of our supporters, and especially to Eric Schmidt, who has been an extraordinary ally, and to Tom Kalil, Adam Marblestone, and Tony Kulesa, who all helped build FutureHouse into what it is today.