FutureHouse
Eric-Schmidt-backed nonprofit building AI agents to automate scientific discovery — free for researchers.
Overview
FutureHouse is a San Francisco nonprofit research lab, backed by Schmidt Sciences, building an 'AI Scientist' to accelerate fundamental biology and chemistry. Its free platform offers specialised agents — Crow (literature Q&A with citations), Falcon (deep literature synthesis), Owl (prior-work detection) and Robin (end-to-end discovery) — plus an API. In November 2025 it launched Kosmos, its most powerful AI Scientist agent, and spun out the for-profit Edison Scientific to commercialise the technology for pharma and biotech.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 24 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
ASEAN Perspective
FutureHouse in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
What this is for: Automating literature review, hypothesis generation and multi-step scientific reasoning with citation-grounded AI agents.
Who this is for: Academic researchers and scientists in biology, chemistry and adjacent fields.
Availability: Free public platform and API from a nonprofit; heavier commercial workloads are served by its for-profit spinout, Edison Scientific.
What people say
FutureHouse sits in the buzzy "AI Scientist" space and has drawn cautiously optimistic attention. Its earlier agent PaperQA2 was reported to match or beat PhD-level researchers on some literature tasks, and its Robin system made headlines for autonomously proposing an existing glaucoma drug (ripasudil) as a candidate for dry AMD. The November 2025 launch of Kosmos — spun into the for-profit Edison Scientific — drew public praise from figures including Sam Altman, who called it "exciting."
The criticism is pointed and comes from working scientists. Kosmos's own accuracy claim is roughly 79.4% of report statements judged correct by independent reviewers — impressive for an autonomous agent, but far from reliable, meaning roughly one in five statements needs correction. Researchers such as Ben Glocker (Imperial) and Noah Giansiracusa (Bentley) stress that heavy human oversight remains essential and that these agents accelerate rather than replace scientific judgement.
There is no consumer review footprint here — the audience is academic researchers, and sentiment lives in blog analyses and practitioner threads. Treat its outputs as leads to verify, not conclusions to trust. While it is a useful literature-and-hypothesis accelerator with citation grounding, its hallucination rate and self-reported evaluations are significant qualifiers.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including FutureHouse's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with FutureHouse unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to FutureHouse directly →
Spotted something out of date? Suggest an update →
More in Research & Data
