GPT-5 in Healthcare: What Can (and Can’t) It Do?

Want to listen to this post instead? Enjoy the audio version!

The release of ChatGPT-5 by OpenAI is already making noise and, like any industry, healthcare is no exception. The possibilities seem endless: from faster clinical documentation to more natural patient communication.

During the announcement, OpenAI's CEO, Sam Altman, said  "GPT-5 is the best model ever for health… [the new model] scores higher than any previous model on HealthBench, an evaluation that we created with 250 physicians on real-world tasks."

But before we get ahead of ourselves, let’s be clear: GPT-5 does not and will never replace clinicians. And using it in healthcare requires rigorous validation.

This post explores the potential of GPT-5 in healthcare, based on OpenAI’s official documentation and public demonstrations. Since the rollout, many users have expressed skepticism, with some even preferring the previous GPT-4o model. That’s why it’s important to clarify: this post is based on early-stage information. This includes OpenAI’s live demo, official documentation, and evaluations like HealthBench, developed with input from over 250 physicians.

Real-world performance and feedback will shape the real value of this model over time. In the meantime, let's go into the potential of GPT-5 in healthcare based on what experts and OpenAI have shared so far.

What’s New with GPT-5?

Before diving into healthcare use cases, let’s break down what’s changed:

One Unified Model

Unlike previous versions, there’s no longer a need to choose between different GPT versions. You now get a single, more capable model with every use. In a nutshell, it’s simpler and more consistent. However, this change has also been met with criticism: some users report that the new model can produce more errors in certain scenarios, and the inability to switch back to a previous version has sparked debate about flexibility and reliability.

Smarter Voice Mode

The voice experience is on a new level. GPT-5 now responds faster, speaks more fluidly, and even replicates natural human behaviors like breathing, stuttering, and changing intonation. This could be game-changing for virtual care assistants or therapeutic bots.

Better Coding Capabilities

In coding demonstrations, GPT-5 delivered far more sophisticated results than GPT-4, sometimes producing hundreds more lines of optimized code with better context understanding. This matters, for example, when integrating with EHRs or building medtech apps.

Improved Memory

GPT-5 now remembers context over much more extended conversations. That’s especially valuable in healthcare, where continuity and patient history are key.

Faster Responses

OpenAI’s team noted during this new model’s presentation that it processes outputs nearly twice as fast as previous versions, reducing friction in administrative workflows.

Copywriting & Communication

Whether it’s writing a summary, marketing copy, or patient-friendly content, GPT-5 offers stronger, more creative expression and tone control. Setting the assistant’s personality (e.g., warm, direct, humorous), customization is now a native feature.

Personalization

The new accent color customization in GPT-5 may seem minor. Still, it enhances the user experience by allowing interfaces to align with brand identity, improve emotional tone (e.g., calm or energizing), and support accessibility.

Chat GPT-5 Pricing model

While GPT-5 is available to free users, access is limited. You can use up to 10 prompts every 5 hours. So, it’s not entirely unrestricted.

Health-Specific Improvements

In the GPT-5 announcement, Altman revealed that one of the most common ways people use ChatGPT is to check information about their health. This insight prompted the company to strengthen the model’s healthcare capabilities in the new release.

GPT-5 is better at:

• Interpreting lab results and medical studies
• Offering non-diagnostic health suggestions
• Understanding medical terminology with more subtlety 

The Healthcare Promise and the Responsibility: two sides of the same coin.

So, what does this mean for healthcare professionals and product teams?

GPT-5 can:

• Accelerate clinical documentation
• Support health education 
• Power smarter chatbots for triage or onboarding
• Help developers build safer, faster, and more personalized health tools

Altman made it clear: health is one of the areas where this technology can really make a difference. According to the founder and CEO, GPT-5 will "help you understand your healthcare and make decisions on your journey." He made a bold claim, "anyone, pretty soon, will be able to do more than anyone in history could."

But let's remember something very important:  This tool is not an FDA-cleared tool, and it is not necessarily HIPAA compliant. It doesn’t replace human judgment. And any use in clinical settings must go through proper validation processes.

Just because it “sounds right” doesn’t mean it’s clinically safe. Even with stronger guardrails, LLMs can still hallucinate. That’s why the real innovation lies in how teams integrate, monitor, and verify these tools.

Final Thoughts

ChatGPT-5 represents an exciting step forward in performance, customization, and speed. It opens up new possibilities in digital health and software development. But the reality is: it’s still early. Also, as stated pre: GPT-5 is a tool, not a doctor. Validation is critical. Oversight is non-negotiable.

The best way forward? Explore, test, and co-create responsibly. Like any powerful tool, its real value will depend on how it's implemented in real-world healthcare settings.

If you're navigating how this fits into your product or workflow, we’re here to help build what’s next in healthcare, together.