Question 1

How much funding did healthcare AI startups raise in 2025?

Accepted Answer

Healthcare AI startups raised approximately $18.2 billion in venture capital funding in 2025, according to Rock Health's annual digital health funding report. This represents a 34% increase over 2024 and makes healthcare AI the single largest category of AI venture investment outside of foundation model companies. The funding was concentrated in a few large rounds: the top 10 deals accounted for $9.1 billion, or roughly half of all healthcare AI investment. Key recipients included Tempus AI ($2.1B), Hippocratic AI ($850M), and Recursion Pharmaceuticals ($700M).

Question 2

How many AI medical devices has the FDA approved?

Accepted Answer

The FDA authorized 12 new AI/ML-enabled medical devices through its 510(k), De Novo, and PMA pathways in 2025 that involved genuinely novel AI capabilities. However, this number requires context: the FDA's total count of 'AI-authorized devices' is higher (approximately 950 cumulative through 2025) because it includes iterative updates to previously authorized devices and products where AI is a minor component. The 12 figure represents truly new AI products that reached market for the first time with autonomous or semi-autonomous clinical capabilities.

Question 3

Why is FDA approval for AI so slow?

Accepted Answer

FDA approval for AI medical devices is slow for three structural reasons. First, the FDA's regulatory framework was designed for static medical devices, not software that updates continuously — the agency is still developing its approach to 'predetermined change control plans' that would allow AI to improve post-approval. Second, clinical validation for AI requires prospective studies demonstrating that the AI performs as well or better than standard of care, which takes 18-36 months minimum. Third, the FDA has approximately 120 reviewers qualified to evaluate AI/ML submissions, handling roughly 300-400 submissions per year, creating a structural review bottleneck.

Question 4

Which healthcare AI companies are actually generating revenue in 2026?

Accepted Answer

The healthcare AI companies generating meaningful revenue fall into three categories. First, diagnostic AI companies that received FDA clearance before 2024: Viz.ai (stroke detection, ~$120M ARR), Aidoc (radiology triage, ~$80M ARR), and Caption Health (cardiac ultrasound, acquired by GE). Second, clinical documentation AI companies that avoid FDA regulation: Abridge (~$100M ARR), Nuance DAX (Microsoft, ~$500M+ ARR), and Nabla (~$45M ARR). Third, drug discovery AI platforms selling services to pharma: Recursion (~$180M revenue) and Tempus (~$600M revenue, though most from diagnostics lab services rather than AI). The companies with the largest funding rounds are generally not the ones with the most revenue.

Question 5

Is healthcare AI a bubble in 2026?

Accepted Answer

By traditional venture metrics, healthcare AI shows bubble characteristics: the median pre-revenue healthcare AI startup raised at a $400M+ valuation in 2025, the funding-to-revenue ratio across the sector is approximately 14x (compared to 4-6x for enterprise SaaS), and the time-to-revenue for FDA-regulated products is 4-7 years from founding. However, the long-term opportunity is real — the US healthcare system generates $4.5 trillion in annual spending with enormous inefficiencies that AI can address. The question is not whether healthcare AI is valuable but whether current valuations accurately reflect the 5-10 year timeline required to capture that value through the regulatory process.

Question 6

What is the FDA's approach to regulating AI in healthcare?

Accepted Answer

The FDA has been developing a regulatory framework for AI/ML-based Software as a Medical Device (SaMD) since 2019. The current approach includes three key elements: a risk-based classification system that applies different review standards based on the clinical risk of the AI's decisions, a 'predetermined change control plan' (PCCP) framework that allows AI developers to define in advance how their algorithms will change post-market, and a real-world performance monitoring requirement. In 2025, the FDA also established a dedicated Center for AI in Medical Devices with a $120M annual budget, signaling increased regulatory capacity but not yet matching the pace of industry submissions.