The Q1 2026 rush: Where AI giants forayed into healthcare

Every AI company wants to win health. Health is one of the hottest AI markets right now.

That’s how Leonard Rinser summed it up, and Q1 2026 proved him right.

For years, Big Tech has been steadily building in healthcare.

From electronic health records and cloud infrastructure to wearables, AI research, and hospital partnerships, companies like Google, Microsoft, Apple, and Amazon have all been deeply embedded in different parts of the system.

But these efforts were fragmented. Different bets. Different timelines. Different layers of the stack.

Then, in the first three months of 2026, those parallel efforts started to converge.

• OpenAI launched ChatGPT Health.
• Anthropic followed with Claude for healthcare.
• Google connected its research with consumer health data.
• Microsoft, Amazon, and others rolled out integrated AI layers.

One after the other. Almost like we’d seen before with ChatGPT, Gemini, Copilot, Claude, Perplexity.

But this time, the stakes are different.

Because this isn’t search or productivity. It is HEALTHCARE. Stricter regulations, sensitive personal data, and far-reaching consequences.

Will these AI giants survive the healthcare game? That’s something we’ll be watching closely.

For now, let’s look into how these giants are foraying into healthcare.

OpenAI in healthcare

January 2026. OpenAI made the first move. Launched ChatGPT Health, followed by OpenAI in healthcare.

ChatGPT Health

ChatGPT Health turns a generic AI assistant into something more personal and more contextual.

Through partnerships like b.well, users can:

• Connect medical records from U.S. providers.
• Sync data from apps like Apple Health, Function, and MyFitnessPal
• Interpret lab results
• Prepare for doctor visits.
• Track their health patterns.

The company is positioning AI as a day-to-day health companion. Emphasising privacy, OpenAI assures that health conversations will stay isolated and not be used to train foundation models.

OpenAI for healthcare

Just the next day, OpenAI launched its enterprise stack: OpenAI for Healthcare.

• Secure, centralised workspace
• Role-based access controls
• Single sign-on and governance features
• Designed for hospitals, providers, and health systems

This focuses on evidence-based reasoning, reducing administrative burden, supporting clinical workflows, and bringing AI into the system.

Horizon1000: $50 Mn partnership with Gates Foundation

Then, OpenAI announced a $50 million partnership with the Bill & Melinda Gates Foundation, aimed at deploying AI across 1,000 clinics in Africa by 2028, starting with Rwanda.

The initiative, named Horizon1000, is one of the largest AI-for-health commitments on the continent to date. It will combine funding, technology, and technical support from the Gates Foundation and OpenAI.

Sub-Saharan Africa is estimated to face a shortage of nearly six million healthcare workers. The systems there are already overstretched, and the gap cannot be closed through training alone.

AI, here, will become infrastructure. It will give OpenAI’s models the real-world validation it requires to be replicated elsewhere.

ChatGPT for Clinicians

To foray into clinical workflows, OpenAI has now introduced ChatGPT for Clinicians, a version tailored for healthcare professionals.

It is being made free for verified clinicians (physicians, NPs, PAs, pharmacists) in the U.S. ChatGPT for Clinicians supports:

• Clinical documentation and note-taking
• Medical research and evidence synthesis
• Care consult and patient communication

Already, millions of clinicians globally are using ChatGPT weekly, with adoption rapidly increasing across healthcare systems.

Now, OpenAI wants its AI to move from general AI assistant to a clinical co-pilot embedded in everyday care.

GPT Rosalind

Earlier this month, OpenAI expanded beyond care delivery with the launch of GPT-Rosalind, a domain-specific model for life sciences and drug discovery.

Designed for scientific workflows, it enables:

• Literature synthesis across research papers
• Hypothesis generation
• Experimental planning
• Multi-step biological reasoning

What makes this move significant is early real-world adoption. GPT-Rosalind is already being piloted by major pharma and biotech players like Amgen, Moderna, Thermo Fisher Scientific and Novo Nordisk.

These companies are applying the model across R&D workflows, signalling that OpenAI is not just experimenting, but embedding itself into the drug discovery pipeline.

Anthropic in healthcare

Next to the foray was Anthropic. If OpenAI led with the consumer story, Anthropic did almost the opposite.

Claude for healthcare

Anthropic launched a HIPAA-ready healthcare suite designed for prior authorisations, care coordination and regulatory documentation.

Built for healthcare’s complexity, the tech giant embedded connectors to stitch the fragmented health system.

It started with:

• CMS coverage databases
• ICD-10 coding systems
• NPI registries
• PubMed

And extended capabilities into:

• FHIR development
• Clinical trial workflows
• Regulatory systems like Medidata

The Coefficient Bio acquisition

Next, big news from Anthropic came earlier this month, when it acquired Coefficient Bio for ~$400 million.

On the surface, it looked surprising. A relatively young startup (under a year old) commanding that kind of valuation.

But what Coefficient Bio brings explains the premium. The startup operates at the intersection of computational biology, wet lab experimentation and AI-driven drug discovery.

By combining Claude’s reasoning capabilities with Coefficient Bio’s infrastructure, Anthropic can:

• Accelerate drug discovery cycles
• Improve clinical trial design
• Build closed-loop systems where:
  • AI suggests hypotheses
  • Experiments validate them
  • Results feed back into the model

And that is the holy grail in life sciences: A continuous feedback loop between AI models and real-world biology. Anthropic is playing big.

Google

In the first quarter, we saw Google moving from building medical AI to owning the entire health journey.

MedGemma + MedASR: The intelligence layer

Google continued advancing its medical AI stack with:

• MedGemma 1.5
  • Supports CT, MRI, histopathology
  • Enables longitudinal analysis—comparing scans over time, not just snapshots
  • Understands lab reports and anatomical context
• MedASR
  • Medical speech-to-text model
  • Built for noisy, real-world clinical environments
  • Improves dictation accuracy in complex settings

This isn’t only “AI for hospitals.” It’s the developer infrastructure for health AI applications at scale.

Furthermore, Google announced the MedGemma Impact Challenge on Kaggle, with $100,000 in prizes for developers exploring new ways AI can advance healthcare and life sciences.

Fitbit + Health Connect: The consumer health layer

Then, at the Google CheckUp 2026, the giant announced that Fitbit users can:

• Connect medical records
• Sync wearable data (heart rate, sleep, activity, SpO₂)
• Integrate lab results, medications, and CGM data
• Bring fragmented health data into one place

And with over 30 million Fitbit users, Google upgraded its existing user base into a health data platform.

Gemini-powered insights

On top of this sits Google’s AI layer:

• Gemini acts as a personal health reasoning engine
• Interprets patterns across time and data sources
• Moves from raw metrics to meaningful insights

Not just: “your sleep dropped last night”

But: “your sleep has declined over 3 months alongside rising resting heart rate. Here’s what that could mean”

From insight to care: Google is also closing the loop

Beyond data and insights, Google is actively building pathways into care:

• Partnerships with CVS Health to bring AI-driven decisions closer to real-world care.
• Integration with clinical systems and care navigation platforms
• Layering virtual care and provider networks on top of the data ecosystem

When we put it all together, as Leonard Rinser says, Google is building the full stack:

• Wearable hardware (Fitbit + Pixel Watch)
• Real-time biometric data (heart rate, sleep, activity, SpO2)
• Clinical validation (Nature-published metabolic research)
• Medical records integration (lab results, medications, visit history)
• CGM data (continuous glucose monitoring via Health Connect)
• AI reasoning layer (Gemini-powered Personal Health Coach)
• Care navigation (Included Health partnership for virtual visits)

Google is positioning itself as an infrastructure for continuous, connected care.

Amazon

Amazon didn’t make a flashy entry like others. It remained calm but announced AI updates to keep up with the competition.

One Medical Health AI assistant

Inside its primary care platform, One Medical, Amazon introduced:

• AI-driven patient interaction
• Health navigation support
• Guidance embedded into care delivery

AWS: Amazon Connect Health (AI agents)

On the infrastructure side, AWS launched:

• AI agents for healthcare providers
• Automated clinical notes
• Scheduling, billing, and workflow automation

This directly targets one of healthcare’s biggest problems: Administrative overload.

Amazon positioned itself as the operational backbone of healthcare.

Microsoft

Microsoft played to its biggest strength: distribution. It launched Copilot Health.

Copilot Health

Microsoft’s healthcare Copilot connects:

• 50+ wearables (ŌURA, Fitbit, etc.)
• Records from 50,000+ hospitals via HealthEx
• Lab data from Function Health.

It turns fragmented health data into a coherent, longitudinal narrative.

The giant has guaranteed strict privacy control, with Copilot Health data staying separate from general Copilot and will not be used for model training.

Partnership with Bristol Myers Squibb

Next, Microsoft partnered with Bristol Myers Squibb to speed up lung cancer detection.

BMS will deploy FDA-cleared radiology AI algorithms through Microsoft’s Precision Imaging Network to automatically analyze X-ray and CT scans and help surface hard-to-detect lung nodules, with the goal of identifying some patients earlier and triaging them into the right care pathway.

What makes Microsoft’s position unique is the scale. Microsoft’s network is already used by more than 80% of U.S. hospitals.

And that’s the distribution most players can’t match.

Apple

While all tech giants made loud announcements, Apple quietly made a fundamental move.

FHIR R4 integration

Apple upgraded its Health app to support FHIR R4, aligning with modern EHR standards.

This enables:

• Seamless data exchange with clinical systems
• Continuous data access (not episodic uploads)
• Stronger remote monitoring and care programs

On the other hand, Apple has already established itself in medical research for brain-computer interface (BCI) with its BCI HID framework.

Apple is blurring the boundary between consumer data and clinical records.

Perplexity

Last but not least, Perplexity foraying into health with Perplexity Health. It entered as an AI-native challenger.

Perplexity Health

It connects:

• Apple Health
• EHRs across 1.7 Mn+ providers
• Wearables like Fitbit, Ultrahuman, Withings (and Oura coming soon)

The AI health agent doesn’t just find links, it queries the body’s data in real time to give contextual, personalised answers.

What makes Perplexity Health different is its decision to add a human Physician Advisory Board to solve potential inaccuracies.

What experts warn about the tech giant’s AI health race?

LLMs weren’t built for this.

Myoung Cha, Chief Product Officer at Verily, points out that general-purpose LLMs weren’t built for longitudinal health analysis. They lack longitudinal reasoning, clinical context, personal baselines and seasonal patterns.

They can imitate statistical reasoning, sometimes impressively. But the intelligence and reliability are jagged.

He explains how with the same data, LLMs can give widely different answers. And that in healthcare can have serious outcomes.

Data isn’t the same as understanding.

Prashant Palepu highlights that AI can retrieve data from six different lab reports, but it cannot understand your lived context, explain decisions like a doctor or translate numbers into life impact. And that context matters.

Because healthcare is not just data, it’s interpretation.​

The design problem no one is solving yet

Maria Prokhorova made an interesting point. While all the giants launched AI health products, no one showed the interface.

ChatGPT Health. Copilot Health. Perplexity Health. All say “Join the waitlist”. Nobody has shown the screen where a real person looks at their health data and decides what to do next.

And in healthcare, that’s important because too much data can overwhelm users, resulting in low engagement. Insights are not the same as usability.

She argues that the real winner here may not be the best model. It may be the one who gives the best experience.

With health, the risk is bigger.

Leonard Rinser puts it bluntly. He says,

Healthcare doesn’t follow the “move fast” playbook. Here, speed without safety is reckless.

So, is it all negative?

No, not all is negative, there are positive aspects too:

They have the scale to make a difference.

Dr Franz, CEO at deepc, points out that generalist AI companies entering healthcare was the biggest unseen force for the healthtech space. Though obvious.

“These companies sit at the frontier of AI, operate with orders of magnitude more compute, and attract >100x the venture capital. Their scale can push the state of the art faster than anyone else.”

They are addressing a big gap.

Petya Kamburova points out that these platforms are not trying to replace doctors. They’re giving context that the doctor doesn’t have the time to provide.

“Most appointments last 30 minutes. Most cardiologists are not following how your liver or brain performs. Your biology is a holistic system generating signals 24 hours a day. That gap is where these tools live. They have the ability to spot a trend in biomarkers months before a symptom ever appears.”

One open question: where are the regulators?

While innovation is accelerating, regulation feels slower.

• No unified global framework
• Limited clarity on validation standards
• Data governance is still fragmented

There needs to be guardrails in place before AI health tools go live.

Wrapping up

Q1 2026 didn’t introduce AI to healthcare. It marked the moment when every major tech player decided they couldn’t stay out anymore. Healthcare became the next battleground for AI platforms.

As Guido Giunti pointed out, what’s interesting is that they’re all betting on different entry points:

• personal health companion
• evidence-first search
• workflow infrastructure
• enterprise / life sciences collaboration

They are all different roads to the same destination. And the race is on.

But in healthcare, winning won’t come from launching first. It will come from building trust and proving it over time.

Who do you think among these giants will actually make a difference?

-By Jhanvi Shah and the AHT Team