In partnership with
AIHealthTech Insider: Issue # 112
August 3, 2026
This week: the app millions use for health questions can now connect to your medical records. Anthropic is putting Claude to work on rare diseases. And scientists found a hidden flaw in the AI tools racing to design new drugs.
Here's what changed, in plain English.
Summaries are for education, not medical advice. Always verify locally before clinical use.
☕ Quick question before we dive in 👇
Image: AI-generated, AIHealthTech Insider
Be honest — how many cups of coffee a day? (There's a reason we're asking. Keep reading.)
🔬 The Big Story
ChatGPT Can Now Connect to Your Medical Records and Apple Health
OpenAI is rolling out Health in ChatGPT for U.S. users. Turn it on, and ChatGPT can connect to your Apple Health data and supported medical records — lab results, medications, past visits, sleep, activity — and use them to answer you. It can explain a lab number, show how a result changed, or help you prep for an appointment. OpenAI says 300 million+ people already ask it health questions every week.
Image: AI-generated, AIHealthTech Insider
Why it matters: your health info usually lives in a dozen scattered places. Pulling it into one spot you can talk to is genuinely useful — and you control what you share.
But here's the honest part: health information stored in a consumer AI service may not get the same legal protections it had inside your doctor's system. HIPAA generally covers healthcare providers, insurers, and their business partners — not every consumer app. OpenAI says connected records are encrypted, aren't used to train its models or target ads, and can be disconnected anytime. It also warns ChatGPT can make mistakes and doesn't replace professional care. Bottom line: understand the privacy settings before connecting anything sensitive.
👉 Official source: https://openai.com/index/health-in-chatgpt/
New here? This is AIHealthTech Insider — AI healthcare signals every week, plain English, no jargon. Subscribe free below.
Sponsored
Talk to your AI tools the way you'd talk to a colleague.
You don't send a colleague a three-word brief. You explain the context, the constraints, what you've already tried. But typing all that into ChatGPT takes forever — so you don't.
Wispr Flow lets you speak your prompts instead. Talk through your thinking naturally and get clean, paste-ready text. No filler words. No cleanup. Just detailed prompts that actually get you useful answers on the first try.
Millions of users worldwide. Works system-wide on Mac, Windows, and iPhone.
⚡ Quick Hits
🔸FDA Clears an AI That Helps Read Breast Ultrasounds
Image: AI-generated, AIHealthTech Insider
DeepHealth, part of RadNet, received FDA clearance for AI that helps read breast ultrasounds—detecting areas of concern, describing them, and generating reports. The goal is to improve consistency and help radiologists work more efficiently.
🔸Claude Is Being Put to Work on Rare Diseases
Image: AI-generated, AIHealthTech Insider
Anthropic created a program supporting scientists and early-stage biotech teams studying rare genetic diseases. Selected teams receive up to $50,000 in Claude credits to analyze data, identify biological mechanisms, and support early drug development. Outputs from the basic-science track will be shared publicly.
🔸Scientists Found a Major Flaw in Protein-Predicting AI
Image: AI-generated, AIHealthTech Insider
Researchers found that leading protein-prediction models sometimes produced physically implausible structures, particularly when engineered sequences included charged residues in unusual locations. Their takeaway: validate AI predictions with physics-based simulations before using them to guide experiments or drug research.
Sponsored
Your employees are connecting AI to everything. Now what?
ChatGPT and Claude aren't just answering questions. Employees are connecting them directly to Notion, Linear, Jira, and the rest of your stack — with no security visibility into what data moves or what actions they take.
Harmonic Security gives your team the visibility to control it.
Know someone in healthcare, medicine, or AI? Forward this to them. They can subscribe free here →
🌍 Beyond AI
You've Been Told to Eat More Protein. A Big New Review Says: Maybe Not.
A major new scientific review challenges the idea that more protein is always better. Across decades of studies, eating less protein — and less of a few building blocks like methionine and isoleucine — was linked to better metabolic health and, in animals, longer lives.
Image: AI-generated, AIHealthTech Insider
The catch: the strongest evidence is in mice and flies, not people — and cutting protein isn't for everyone. Older adults at risk of muscle loss, pregnant women, kids, and anyone recovering from illness generally need more, not less. This isn't "stop eating protein"; it's that "more is always better" may be oversold.
Sponsored
One idea shouldn't take six rewrites to post.
Posting everywhere means rewriting one idea six times, so you post to one, or none. SureThing turns one idea into native posts for every platform.
💡 Good to Know — About That Coffee...
Here's why we asked. A new American Heart Association review found that for most healthy adults, up to about 400 mg of caffeine a day — roughly three to five 8-ounce cups — is generally safe for your heart, maybe even helpful. (Many coffee-shop drinks are far bigger and stronger than 8 ounces.)
The catch: this applies mainly to regular coffee—not highly caffeinated energy drinks and shots. Their larger caffeine doses and additional ingredients may increase blood pressure or trigger abnormal heart rhythms.
Stay ahead: AI Healthcare News!
This is just a glimpse of the AI healthcare revolution. Subscribe for exclusive updates on medicine’s future!
