In partnership with
AIHealthTech Insider: Issue # 98
April 27, 2026
AI cracked the blood-brain barrier for Alzheimer's, a chatbot became a teenager's most dangerous confidant, Nature Medicine told the industry to prove AI actually helps and a 200,000-person study found the most powerful health intervention costs nothing.
Summaries are for education, not medical advice. Always verify locally before clinical use.
🔬 The Big Story
AI is transforming healthcare faster than anyone can measure and that’s becoming the problem.
Nature Medicine dropped a bombshell this week. Not a breakthrough. Not a new tool. A question: Is any of this actually making patients healthier?
An April 21 editorial criticized companies and researchers for making bold claims about AI's clinical impact without sufficient evidence. A companion paper from the University of Michigan and University of Toronto highlights the lack of supporting evidence for AI tools used in healthcare tasks.
Image source: Grok/AIHealthTech Insider
But here's what nobody's answering:
Does the doctor trust what the AI says? Nobody's tracking that systematically.
Do AI-drafted notes change how physicians think about patients? We genuinely don't know.
When AI speeds up an X-ray read, does the patient actually get better care? The studies haven't been done.
Take AI scribes - the tools that listen to doctor-patient conversations and write up the notes. Clinicians love them. Burnout drops. Paperwork vanishes. But some early research suggests AI tools can alter how people cognitively process information.
Could a scribe subtly reshape what a doctor notices or remembers?
Could it change a diagnosis? We don't have answers yet.
This isn't an anti-AI paper; both authors are believers. However, they emphasize that the industry must shift from proving AI works to proving it helps, highlighting a significant gap between the two.
Sponsored
Last week Viktor wrote a brief, built a landing page, and opened a pull request.
Last week, Viktor wrote a campaign brief, built a landing page, opened a pull request, generated a board-ready PDF from live Stripe data, and sent a follow-up email to a churned customer. All from Slack. Same colleague that also pulls your reports and monitors your dashboards. 5,700+ teams. 3,000+ integrations.
⚡ Quick Hits
Image source: Grok/AIHealthTech Insider
🔸A teen’s AI chatbot ‘friend’ and a wrongful death lawsuit
A 16-year-old named Adam Raine started using ChatGPT for homework then began confiding suicidal thoughts. Court documents allege the chatbot mentioned suicide over 1,200 times. His parents, unaware, are now suing OpenAI for wrongful death. ECRI has named AI chatbot misuse the #1 health technology hazard of 2026. With three-quarters of teens using AI for companionship, the Raine case is forcing an industry-wide reckoning.
👉 Source → The National News Desk
🔸An AI just found Alzheimer's drugs that can actually get into the brain — the part nobody could crack
Most Alzheimer's compounds fail to cross the blood-brain barrier. DeepDrugDiscovery, an open-source AI platform, screens for brain-penetrating compounds that enhance cellular cleanup safely. Two lead compounds crossed the barrier, cleared toxic protein clumps, and restored memory in animal models. The platform is now available for global use.
🔸A simple "gut reset" may solve Ozempic's biggest problem: the weight coming back
About 70% of people who stop GLP-1 drugs like Ozempic regain most weight within 18 months. A new procedure, duodenal mucosal resurfacing, may help. In the REMAIN-1 trial, it reduced weight regain by 40% compared to a sham procedure, with some maintaining over 80% of their original weight loss. A 300-person trial is underway, with results expected later this year, potentially offering a solution for GLP-1 users.
Sponsored
Smarter browsing. Your data never leaves the room.
Most AI tools are a trade — your data for intelligence. Norton Neo breaks that deal. Powerful built-in AI, anti-fingerprinting, VPN, and ad blocking come standard. No setup. No add-ons. No compromises. Search, summarize, and write with AI that works inside your browser and stays there.
🌍 Beyond AI
That thing you won't forgive? It's slowly wrecking your body. Science has the receipts.
This one isn’t about AI. But it’s too important to ignore.
A study of over 200,000 people in 23 countries found that those more inclined to forgive report significantly better psychological well-being a year later. Published in NPJ Mental Health Research, the findings support research showing that grudges lead to chronic stress, elevated cortisol, higher blood pressure, worse sleep, and increased heart disease risk.
Image source: Grok/AIHealthTech Insider
The key insight? Forgiveness is a learnable skill. A study in five countries found that a two-week forgiveness workbook reduced depression and anxiety through self-reflection, without apps or prescriptions. Fred Luskin from the Stanford Forgiveness Projects says grudges come from unmet expectations, and forgiveness breaks the cycle without excusing past actions.
Sponsored
Slow down aging at the biological level.
Aramore is a completely new approach to skincare—one that helps your body produce more NAD+, the vital co-enzyme responsible for our cellular health and how we age.
See firmer, more radiant, more resilient skin in just 28 days. Get 20% off your first order with code NEWSLETTER20.
📊 Stat of the Week
81% of U.S. physicians now use AI professionally up from 38% two years ago. (AMA 2026 Physician Survey)
🐰 Rabbit Hole
Three links worth falling into this week:
Image source: Grok/AIHealthTech Insider
Healthcare AI lawsuits are getting wild. Kaiser paid $556M. Cigna's AI denied 300K claims at 1.2 seconds each. Patients are suing over AI scribes capturing health details without consent. → Read it
Your health data is worth billions. Who owns it? A Dutch researcher built a 167,000-person dataset over 30 years. A Chinese AI group downloaded it for €2,000. Nature Medicine investigates the geopolitics of biomedical data. → Read it
Google processes 1 billion health questions a day. At its Check Up 2026 event, Google also committed $10M to retrain clinicians for the AI era — then quietly removed a feature that was giving people misleading health info. → Read it
💼 Who's Hiring
Abridge — AI clinical documentation (Best in KLAS 2025 & 2026, $5.3B valuation). Deploying across Kaiser, Mayo Clinic, Johns Hopkins. Engineering, product, and clinical roles open. → Careers
Hippocratic AI — Patient-facing AI voice agents for chronic care. 1.8M patient calls completed. Hiring across all departments. → Careers
Corti — Healthcare-native APIs for speech-to-text, medical coding, and documentation. 1M+ patients served weekly. → Careers
💡 The One Thing
This issue covered a lot of ground. We want to know what landed hardest for you.
What's your biggest takeaway this week?
Stay ahead: AI Healthcare News!
This is just a glimpse of the AI healthcare revolution. Subscribe for exclusive updates on medicine’s future!
