OpenAI Acquires Torch for $100M to Boost Health AI
OpenAI has acquired Torch for about $100M in equity. The four-person team will join OpenAI as Torch’s “medical memory” supports ChatGPT Health.
TL;DR
OpenAI has acquired Torch, a tiny health-records startup reportedly valued at about $100M in equity, and quickly shut down the standalone app. The four-person team is joining OpenAI to help build ChatGPT Health, using Torch’s “medical memory” tech to pull scattered data—labs, visits, wearables—into one unified view.
OpenAI has acquired Torch, a tiny AI healthcare startup, in a deal reported to be worth about $100 million in equity, and Torch’s four-person team is joining OpenAI while the standalone app has been shut down. Torch’s core idea—creating a “medical memory” that unifies scattered health records—now appears set to power or bolster OpenAI’s ChatGPT Health push.
OpenAI acquires Torch in $100M deal
OpenAI confirmed it has acquired Torch, with reporting from TechCrunch citing a source who said the purchase price was about $100 million in OpenAI equity. Torch was a very small company—just four people—and the acquisition is being widely read as an acquihire focused on the team and its product direction rather than near-term revenue scale.
The Torch app’s independent journey ended quickly after the deal, with reports indicating the product was shut down as its core team moved under OpenAI’s umbrella. OpenAI has said Torch will be brought together with ChatGPT Health, signaling a tighter healthcare strategy inside the ChatGPT ecosystem.
What Torch built: a “medical memory” for AI
Torch positioned itself as a unified layer that pulls medical data from multiple places—hospitals, labs, wearables, portals, and consumer testing—so an AI system can make sense of the full picture. TechCrunch described the product as an app that combined a person’s medical information from many sources and framed the approach as a “medical memory for AI.”
This “context engine” concept targets one of healthcare’s biggest usability problems: patient records are fragmented across systems, making it hard for both people and clinicians to track longitudinal history. By consolidating data into a single view, Torch aimed to make summarization, pattern-spotting, and practical explanations easier for users.
Why a four-person team commanded $100M
Reporting notes the Torch founders previously worked together at Forward, a high-profile primary care startup known for AI-driven clinics, and that shared operating experience in hard healthcare environments likely boosted the team’s value. Forward raised substantial funding and ultimately shut down in late 2024, underscoring how difficult it is to make primary care models work at scale—even with strong technology ambitions.
In that context, OpenAI’s bet looks less like buying a mature app and more like buying concentrated “scar tissue”: domain knowledge, product instincts, and a running start on integrating personal health data into safe consumer workflows. Silicon Republic also reported OpenAI’s intention to merge Torch’s work into ChatGPT Health, suggesting Torch’s architecture and team fit a roadmap that’s already in motion.
How this fits ChatGPT Health
OpenAI has been positioning ChatGPT Health as a dedicated experience where users can upload records and converse to understand results, prep for appointments, and get guidance on lifestyle questions. Silicon Republic reported OpenAI’s commitment that ChatGPT Health will be private—stating that uploaded health data and chats won’t be used to train OpenAI’s foundation models.
Torch’s “unify everything first” approach complements that product direction, because health conversations become far more useful when AI can see labs, meds, diagnoses, and timelines in one place. The acquisition also lands amid broader competition in medical AI tooling, with Silicon Republic noting other major AI labs launching healthcare-focused offerings as well.
