Heidi has acquired UK clinical AI startup AutoMedica and launched two new products to expand its software from clinical documentation into evidence lookup and patient communications.
It introduced Heidi Evidence, an ad-free research tool within its AI Care Partner platform, and Heidi Comms, a system for managing patient communications across calls, bookings, reminders and follow-ups.
Heidi Evidence draws in part on Anthropic's Claude models. It uses large language models to interpret unstructured clinical conversations and medical literature, then returns real-time responses for clinicians during consultations.
Evidence layer
Heidi Evidence provides summaries with citations and verbatim excerpts, allowing clinicians to check sources during care. It can run as a standalone product or alongside Heidi's AI Scribe feature.
Heidi says its platform has supported more than 100 million clinical interactions globally and is used for more than 2.4 million consultations each week across emergency departments, general practice and specialist clinics. It also says the system supports 110 languages and is used in 190 countries.
Heidi built Heidi Evidence with partners including HealthPathways, EMGuidance, MIMS, Vidal, NICE and BMJ Group. The product focuses on region-specific guidance and formularies.
Heidi is making Heidi Evidence free for individual clinicians, using revenue from enterprise customers to fund access in markets where budgets are constrained or care is fragmented.
Heidi positioned the product as a response to concerns about advertising and influence in consumer AI tools, and as a way to keep clinical research separate from ad-supported search models.
"We believe that for AI to be a true care partner, the integrity of its evidence must be non-negotiable," said Dr. Thomas Kelly, Co-Founder and CEO of Heidi.
"As we see more general-purpose AI platforms like OpenAI move toward ad-supported models, consumers are rightly concerned about hidden influence.
In a healthcare setting, that concern becomes paramount. Bringing transparent, clinical-grade insights into the room makes it easier to deliver quality care, but that information must be free from the ambiguity of commercial influence. By committing to Evidence being ad-free and independent, we ensure clinicians can stay present with their patients, knowing their decision-making is built on pure clinical rigor, not a business model."
UK acquisition
AutoMedica developed a point-of-care clinical decision support product focused on UK-sourced content. The startup took part in the MHRA AI Regulatory Airlock Programme, a sandbox for testing and scrutiny of healthcare AI products.
Heidi said the acquisition brings experience from the NHS AI Airlock into its product development, along with regulatory relationships and an evidence-led approach to clinical decision support.
AutoMedica's founders are clinicians with research backgrounds in AI and machine learning. Its team will join Heidi following the deal.
"From the outset we have believed that clinical AI must be built within real healthcare structures, not bolted on around the edges," said Dr Ben Turner, Co-Founder of AutoMedica.
"Working through the NHS AI Airlock has shown that when regulators, clinicians and technologists come together, it is possible to build AI that is both ambitious and safe by design. Joining Heidi means we can take what we have learned and apply it at a far greater scale, within a platform that shares our focus on evidence, safety and the realities of day-to-day clinical practice."
Product expansion
Heidi is broadening its scope beyond transcription and note generation. Heidi Comms extends the platform into administrative tasks around appointments and follow-up, which Heidi describes as an AI partner for healthcare teams.
Heidi is also adding clinical reasoning and evidence retrieval to its workflow. It says clinicians face a growing volume of medical research and guidance, making it difficult to stay current through traditional methods.
Heidi Evidence is positioned as a research function within the same interface as documentation.
Heidi's use of Anthropic's models places it among healthcare AI vendors that select third-party foundation models for specific tasks. Heidi said it chose Claude for handling complex conversations and dense literature, and for producing grounded outputs.
Michael Tolo, General Partner at Blackbird, said the approach addresses trust and commercial influence in clinical AI systems.
"Heidi is tackling one of the hardest problems in healthcare AI: how to scale capability without compromising trust," Tolo said. "By treating evidence as core infrastructure, not content monetised through ads or influence, Heidi is building the kind of defensible, globally relevant platform healthcare systems are demanding."
Heidi said Heidi Evidence will launch in private clinical settings in the UK, with an NHS rollout expected to follow.