A good AI answer is not proof of a dependable tool. In health, validation has to be specific: what job is the tool doing, for whom, with what data, under what conditions, and what happens when it fails?

Psilocybin may feel newly mainstream. However, the research questions are older. Roland Griffiths’ TEDMED Talk examines supported conditions, subjective experience, and what researchers observed beyond the session itself. Watch the full Talk: tedmed.com/talk/the-science-…

Roland Griffiths’ TEDMED Talk shows how psilocybin research was already asking about risk, supported conditions, and subjective experience years before today’s surge in public attention. Watch the full talk: tedmed.com/talk/the-science-…

The molecule is only the beginning. Psilocybin’s effects can depend on the person, the dose, the expectations, and the setting around it. TEDMED Clearly breaks down what the compound is and why context matters. Watch TEDMED Clearly: What is psilocybin? youtu.be/JOaX_HfyIYg

Before drawing conclusions about psilocybin’s promise or risks, it helps to start with the basics: what the compound does, how its effects can vary, and the conditions researchers study around it.

Science needs rules. It also needs the capacity to recognize when those rules are no longer serving the people they're meant to help. Revisit Anthony Fauci's original TEDMED Talk on the experience that changed how he thought about listening, evidence, and responding to urgent human need. Watch here: tedmed.com/talk/embracing-fl…

Ending an outbreak and ending its consequences are not the same thing. Soka Moses asks what "recovery" means when survivors are still navigating stigma, barriers to care, and the lasting effects of a crisis. Watch the full Talk: tedmed.com/talk/for-survivor…

Scientific rigor isn't the same as rigidity. We revisit Anthony Fauci's reflections on the HIV/AIDS crisis and what happens when listening to the people most affected changes how science responds. Watch In Context: tedmed.com/talk/embracing-fl…

Good science has to respond to new evidence and changing circumstances. In a health crisis, the challenge is adapting without losing the rigor that makes science trustworthy. This week at TEDMED, we’re looking at what happens when evidence, urgency, and human need are all changing at once.

What if telemedicine’s goal shouldn’t be to imitate an in-person exam? Elliot Swart asks how technology might extend what clinicians can perceive instead. Watch the Talk: tedmed.com/talk/seeing-the-d…

“All the big data” still has to come to bear “one human being to another.” @abe_verghese on what medicine cannot afford to lose as its tools become more powerful. Watch the full Talk: tedmed.com/talk/a-linguistic… @StanfordMed

Patients notice where attention goes. A new TEDMED In Context revisits Abraham Verghese’s ideas about presence in medicine and asks what happens when records, screens, and AI increasingly share the clinical encounter. Watch the new In Context: tedmed.com/talk/a-linguistic… @StanfordMed

Medicine can capture every word and still miss the patient. Better information systems solve one problem. Making someone feel heard is another.

Data can suggest that an existing drug might have an entirely new use. That does not make it a treatment. Prediction finds the possibility. Validation determines whether it is real. Atul Butte on what changes when science learns to ask new questions of data it already has: tedmed.com/talk/what-if-you-…

Biomedical data can do two things at once: Help science discover something new. Reveal something deeply personal about you or people related to you. That tension belongs inside the scientific question, not outside it. @the_yaniv TEDMED Talk shows what that looks like when a family tree becomes a research tool: tedmed.com/talk/how-were-bui…

The next biomedical breakthrough may not begin with collecting more data. It may begin with asking a better question of data science already has. But finding a pattern is only the beginning. Researchers still have to test whether it holds up. Watch our new TEDMED In Context where we revisit Atul Butte's Talk: tedmed.com/talk/what-if-you-…

What does it actually take to be ready for the next outbreak? Testing. Case detection. Clear priorities. Faster response. In his TEDMED Conversation, Ed Kelley looks at what public health learned from past outbreaks and what still has to work better next time. Watch the full Conversation: tedmed.com/conversations-pod…

Vaccines are one part of outbreak preparedness. So are testing, case detection, public-health capacity, and the ability to respond quickly when cases appear. Having the tool is not the same as having the system ready to use it.

The treatment existed. The donor almost didn’t. Seun Adebiyi’s leukemia story is a reminder that medical innovation means little if the infrastructure connecting patients to it is missing. Watch Seun Adebiyi's full talk: tedmed.com/talk/insights-fro…