Part 2: AI is rewriting medicine

I’m serially amazed and hopeful about AI, but I’m not delusional.

For every breakthrough headline, there are real hurdles to using AI to extend human lifespan: regulation, clinical trials, adoption, and the simple fact that scientific progress rarely moves in a straight line (historically, nearly 90% of drugs that enter human trials never reach approval, which is the baseline AI is trying to improve).

This is part 2 of AI is rewriting medicine, and the one question that keeps pulling me back—the reason I created this series—is: “Will any of this happen fast enough to matter for people alive today?”

In Part 1, I explored five ways AI is already changing medicine. Better cancer detection, faster drug discovery, and new tools for understanding diseases like Alzheimer’s are happening now.

My takeaway is that AI isn't coming to healthcare—it’s already here.

Even innovations still in research labs are showing real promise to extend lifespan and healthspan—the years we remain active, independent, and mentally sharp—in guys like me. It might not be perfect, but health never is. I just want to feel good.

This week, let’s look at five more AI frontiers that could have the biggest impact on longevity in the years ahead.

As the body ages, some cells stop dividing but don’t die. They’re no longer functional, but their limbo existence still causes damage. In older tissues, senescent cells can make up 10–15% of total cells, especially in fat, skin, and immune compartments.

Researchers call them senescent cells (AKA ‘zombie cells’). They release inflammatory signals that accelerate the breakdown of surrounding healthy tissue. A joint study from Harvard, Mayo Clinic, and Cedars-Sinai tested a combination of two existing drugs—dasatinib and quercetin—in older adults and found reduced inflammation and improved memory.

What could this mean for you: If you are over 60, zombie cell accumulation is real and potentially dangerous. This is not a distant problem. Formal clinical trials are active at the Mayo Clinic. An approved senolytic protocol—likely an intermittent treatment—is probably within five to seven years. AI is already being used to screen thousands of compounds to identify safer senolytics with fewer side effects. Find a physician who stays current on this science.

Two people can receive the same diagnosis—breast cancer, heart failure, Type 2 diabetes—and have almost nothing biologically in common. The same drug that works well for one person may do nothing for another. Response rates for many common drugs still hover around 50–70%, meaning a large fraction of patients get limited benefit.

Until recently, treatment was often trial and error. Now AI can match patients to therapies based on genetics, metabolism, and even microbiome subtype.

Getting a drug from the lab to the pharmacy has historically taken about 15 years and cost more than $2 billion.

AI is beginning to compress that timeline. New platforms can model disease targets, simulate drug interactions, and predict side effects before a compound is even physically created. In 2023, the first AI-designed drug candidates entered human trials in under 30 months—about one-third of the traditional timeline.

What this means for you: Drugs that once took 20 years to reach patients could arrive in 8–10 years with AI. For adults in their 50s and 60s, treatments currently in development for age-related diseases may realistically arrive within your lifetime window. They’re not here yet, but the timeline is shifting in a meaningful way.

One of the most powerful shifts in medicine is catching disease early and monitoring it continuously instead of relying on annual checkups.

AI trained on longitudinal MRI data can now track brain aging and flag disease progression before symptoms appear. Some models can detect Alzheimer ’s-related changes 5–10 years before clinical diagnosis.

At the same time, the National Institute on Aging is funding programs that combine electronic health records, wearable data (heart rate variability, temperature, respiratory rate), and clinical measurements into predictive models of disease risk.

What this means for you: Wearables already generate the kind of continuous data these systems need. Clinical-grade AI interpretation of that data is likely within three to five years. Meanwhile, multi-cancer early detection tests—like Grail’s Galleri, which screens for signals from more than 50 cancers in a single blood draw—are available now by prescription, with specificity above 99% (though sensitivity varies). For anyone over 45, this may be one of the most accessible forms of AI-assisted early detection today.

In many high-income countries, people spend 15–20% of their lives in poor health. That’s the gap AI is trying to shrink. The goal is not just living longer, but staying functional longer—compressing serious illness into a shorter period at the end of life instead of stretching it over a decade. AI won’t replace biology, but it will accelerate the cycle of hypothesis, testing, and discovery. Some estimates suggest it can reduce early-stage research timelines by 30–50%.

What this means for you: For adults between 50 and 70, the interventions that are most likely to deliver real benefit soon are AI-guided early detection (available now), multi-cancer blood screening (available now), precision medicine matching (already underway), and senolytic therapies (within five to seven years). More experimental areas like epigenetic reprogramming are likely a decade away, but still relevant for those in their 40s.

Maybe the most useful step you can take away from all this is to establish your biological baseline—whether that’s using genome, microbiome, and wearable data. That way, when the next wave of AI tools comes, they’ll be meaningful for you.

The information provided about wellness and health is for general informational and educational purposes only. We are not licensed medical professionals, and the content here should not be considered medical advice. Talk to a doctor before trying any of these suggestions.