LEVELS – A Whole New Level

Medical AI is already getting good at doing things doctors do. Dr. Ziad Obermeyer thinks the bigger opportunity is using AI to discover things medicine doesn’t yet know.

His work shows both sides of that future: algorithms can amplify bad assumptions when trained on the wrong targets, but they can also uncover signals in medical data that humans miss. Obermeyer argues that as more health data is collected outside the hospital, AI could turn it into a continuous picture of health rather than a series of isolated snapshots.

Free course: Improve your metabolic health
Get our free email course on how glucose, nutrition, exercise, sleep, and measurement can help you build habits that support better energy and long-term health: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://levels.link/wnl⁠

What We Cover:
Why AI should learn from patients and outcomes, not just doctors
How an ECG model found hidden risk of sudden cardiac death
Why medical data access is still a major bottleneck
How more measurement could actually mean fewer unnecessary tests
Why healthcare may be entering a “mainframe to PC” transition

🎙️ About the Guest:
Dr. Ziad Obermeyer is an emergency medicine physician and an Associate Professor at the UC Berkeley School of Public Health. He also co-founded Dandelion Health and the non-profit Nightingale Open Science.

📍What Dr. Ziad Obermeyer & Mike Haney discussed:
02:43 Why better data is fundamental to medical AI
05:54 The problem: There’s no variable called “get sick”
09:27 Algorithms optimize exactly what you tell them to
23:03 Why teaching AI to copy doctors limits what it can discover
26:28 How AI could create a new science of medicine
28:40 Could AI predict sudden cardiac death before it happens?
34:14 Can an algorithm teach us what it sees?
36:50 Medicine’s biggest AI bottleneck: access to data
49:00 Healthcare’s “mainframe to PC” transition
55:08 Why more measurement could actually mean fewer tests
58:46 Why medical AI is aiming too low
1:02:13 What the next generation of wearables needs to measure

🔗 Helpful Links:
Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations (Science, 2019)https://www.science.org/doi/10.1126/science.aax2342
An Algorithmic Approach to Reducing Unexplained Pain Disparities in Underserved Populations (Nature Medicine, 2021)https://www.nature.com/articles/s41591-020-01192-7
An ECG Biomarker for Sudden Cardiac Death Discovered with Deep Learning (Nature, 2026)https://www.nature.com/articles/s41586-026-10674-6
Predicting the Future — Big Data, Machine Learning, and Clinical Medicine (NEJM, 2016)https://www.nejm.org/doi/full/10.1056/NEJMp1606181
Nightingale Open Sciencehttps://www.nightingalescience.org/
Dandelion Healthhttps://dandelionhealth.ai/

Watch the conversation: https://youtu.be/WDo-iL6O5nU
Find us on YouTube: ⁠⁠⁠⁠⁠⁠⁠https://youtube.com/levelshealth?sub_confirmation=1⁠⁠⁠⁠⁠⁠

📲 Connect:
Connect with Dr. Ziad Obermeyer on https://ziadobermeyer.com/
👋 Who we are:
Levels helps you understand your metabolic health with personalized data, expert guidance, and tools that connect your daily choices to measurable changes in your body. Our goal is to help you make better decisions about food, exercise, sleep, and long-term health.

Look for new shows every month on A Whole New Level, where we have in-depth conversations with thought leaders about metabolic health.

Why do we crave some foods the more often we eat them? Are we addicted?
Dr. Dana Small’s research suggests that food craving reaches far beyond taste or willpower, and that “addiction” isn’t a helpful descriptor. Every time we eat, unconscious signals from the body teach the brain which foods deliver useful nutrients, and train us to want more of those. In other words, we don’t eat crave foods because they taste good; foods taste good because our body craves them.
Understanding that means we can rethink overeating. It’s not a matter of addiction or self-control; it’s our body behaving the way nature intended it to, but in the setting of foods not found in the wild, like those high in both fat and carbohydrates. But we may also be able to train our brain away from these cravings.

Free course: Improve your metabolic health
Get our free email course on how glucose, nutrition, exercise, sleep, and measurement can help you build habits that support better energy and long-term health: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://levels.link/wnl⁠

What We Cover
Taste vs. flavor, pleasure, motivation, and food reward
How gut nutrient signals teach the brain which foods to seek
Why fat + carbs produce an unusually strong reward response
What eight weeks of high-fat, high-sugar snacks did to the brain
How GLP-1s may cut food wanting while preserving pleasure

🎙️ About the Guest:
Dr. Dana Small is the Canada Excellence Research Chair in Metabolism and the Brain and a professor at McGill University, with appointments in Neurology and Neurosurgery and Medicine. She directs the Modern Diet and Physiology Research Center, and her research examines how sensory, metabolic, and neural signals interact to shape food reward, eating behavior, and metabolic health.

📍What Dr. Dana Small & Mike Haney discussed:
02:44 - Is food actually addictive?
05:02 - Taste and flavor are different senses
08:21 - Why “hedonic” vs. “homeostatic” eating is misleading
10:42 - The reward from sugar begins after you swallow it
22:32 - The “high road” and “low road” of food reward
27:41 - How scientists teach the brain to associate a flavor with calories
38:30 - Why combining fat and carbohydrates produces more reward
47:34 - How beliefs about food can change physiological responses
50:02 - Eight weeks of snacks changes how the brain responds to food
57:51 - What this science means for people trying to eat differently
1:03:50 - How GLP-1 drugs may change food wanting
1:07:30 - Can GLP-1s help retrain eating behavior for the long term?

🔗 Helpful Links:
Habitual Daily Intake of a Sweet and Fatty Snack Modulates Reward Processing in Humans (Cell Metabolism, 2023)https://pubmed.ncbi.nlm.nih.gov/36958330/
Supra-Additive Effects of Combining Fat and Carbohydrate on Food Reward (Cell Metabolism, 2018)https://www.cell.com/cell-metabolism/fulltext/S1550-4131(18)30325-5
Integration of Sweet Taste and Metabolism Determines Carbohydrate Reward (Current Biology, 2017)https://pubmed.ncbi.nlm.nih.gov/28803868/
Rethinking Food Reward (Annual Review of Psychology, 2020)https://pubmed.ncbi.nlm.nih.gov/31561741/
In Defense of Pleasure: We Need to Rethink Food Reward and Obesity (PLOS Biology, 2025)https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3003497
Mind Over Milkshakes: Mindsets, Not Just Nutrients, Determine Ghrelin Response (Health Psychology, 2011)https://pubmed.ncbi.nlm.nih.gov/21574706/

Watch the conversation: https://youtu.be/lXlCqNqjgHY
Find us on YouTube: ⁠⁠⁠⁠⁠⁠https://youtube.com/levelshealth?sub_confirmation=1⁠⁠⁠⁠⁠

👋 Who we are:
Levels helps you understand your metabolic health with personalized data, expert guidance, and tools that connect your daily choices to measurable changes in your body. Our goal is to help you make better decisions about food, exercise, sleep, and long-term health.

Look for new shows every month on A Whole New Level, where we have in-depth conversations with thought leaders about metabolic health.

Twenty-five years ago, sequencing the human genome seemed poised to usher in an era of personalized medicine. Dr. Euan Ashley helped turn that promise into an actual clinical experiment, leading one of the first efforts to interpret an entire human genome in the context of a real patient’s health. The challenge quickly became apparent: medicine could suddenly generate billions of data points about one person, but making sense of them was painfully slow.
That history feels newly relevant. Ashley recently gave an AI system his own 15-year-old genome data and watched it reproduce most of an analysis that had originally required months of work by a team, in about 30 minutes. Meanwhile, genomics is being joined by wearables, multi-omics, and other technologies capable of tracking far more of our biology over time.
In this episode of NextLevel, we use the genome revolution as a case study for what happens when our ability to measure the body advances faster than our ability to interpret and use the information. The next leap in personalized medicine may depend less on collecting another mountain of data than on building the tools, reference datasets, and healthcare systems capable of turning that information into decisions.

Free course: Improve your metabolic health
Get our free email course on how glucose, nutrition, exercise, sleep, and measurement can help you build habits that support better energy and long-term health: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://levels.link/wnl⁠

🎙️ About the Guest:
Dr. Euan Ashley is Chair of the Department of Medicine at Stanford University and the Roger and Joelle Burnell Professor of Genomics and Precision Health, with appointments in medicine, genetics, and biomedical data science. He’s also the author of The Genome Odyssey: Medical Mysteries and the Incredible Quest to Solve Them.
📍What Dr. Euan Ashley & Mike Haney discussed:
03:33 What happened to the promise of personalized medicine?
06:33 The first clinical interpretation of a human genome
11:26 AI interprets in 30 minutes what once took months
21:25 Why genomics still needs better, more diverse reference data
27:41 Building a molecular map of exercise with MoTrPAC
32:01 Exercise changed nearly every organ researchers measured
39:05 Why your personal baseline can matter more than a "normal" range
42:46 How wearables opened a new era of longitudinal health data
47:48 The biggest barrier to continuous, personalized healthcare
51:39 What medical AI needs to become genuinely useful
57:05 How AI can ace medical benchmarks for the wrong reasons
58:20 The heart measurement Ashley considers a medical "holy grail"

🔗 Helpful Links:
Clinical Assessment Incorporating a Personal Genome, The Lancet, 2010. Landmark integrated clinical analysis of a complete human genome.
Molecular Transducers of Physical Activity Consortium (MoTrPAC), NIH Common Fund. Large-scale project mapping how exercise changes tissues and organs. 
MoTrPAC: Human Studies Design and Protocol, Journal of Applied Physiology, 2024. Human study of molecular responses to endurance and resistance training.
The Mitochondrial Multi-Omic Response to Exercise Training Across Rat Tissues, Cell Metabolism, 2024. MoTrPAC study on molecular changes across organs after exercise.
Harmonizing Standards and Resources for the Medical Genome, Nature, 2026. Ashley et al. on genomic reference standards and clinical-grade sequencing.

Watch the conversation: https://youtu.be/bD2YVGdwviE
Find us on YouTube: ⁠⁠⁠⁠⁠https://youtube.com/levelshealth?sub_confirmation=1⁠⁠⁠⁠

👋 Who we are:
Levels helps you understand your metabolic health with personalized data, expert guidance, and tools that connect your daily choices to measurable changes in your body. Our goal is to help you make better decisions about food, exercise, sleep, and long-term health.

Look for new shows every month on A Whole New Level, where we have in-depth conversations with thought leaders about metabolic health.

Is intermittent fasting biohacker wellness hype or legit health intervention? Some studies show real benefits, others find nothing, and the occasional headline calls it bad for your heart.

Dr. Courtney Peterson has spent nearly two decades running some of the field's most tightly controlled feeding trials, and her data points to a clear pattern: there is a real effect to a restricted eating window, but when you eat may matter as much as how long you fast. Her studies found that eating in a short window early in the day lowers blood pressure and improves insulin sensitivity, even without weight loss. A physicist by training, she brings that same rigor to a field built mostly on people's own food diaries.

Free course: Improve your metabolic health
Get our free email course on how glucose, nutrition, exercise, sleep, and measurement can help you build habits that support better energy and long-term health: ⁠⁠⁠⁠⁠⁠⁠⁠⁠https://levels.link/wnl⁠

🎙️ About the Guest:
Dr. Courtney Peterson is an Associate Professor of Nutrition, Harvard T.H. Chan School of Public Health. She holds a PhD in physics from Harvard and leads clinical trials on intermittent fasting, meal timing, and circadian rhythms.

📍What Dr. Courtney Peterson & Mike Haney discussed:
0:00 Introduction
02:39 The three types of intermittent fasting
05:29 The 2012 mouse study that started the field
11:24 Why fasting works through appetite, not extra calories
13:48 Her six-hour study: lower blood pressure, no weight loss
25:34 Why results depend on who's being studied
28:45 The shift worker whose health got worse
36:45 Her theory: timing over fasting length
43:08 Heart attacks, blood pressure, and the body's clock
45:55 The camping trip that reset people's body clocks
58:11 Early findings from her unpublished aging study
1:08:16 Should you eat ice cream for breakfast?

🔗 Helpful Links:
Time-Restricted Feeding without Reducing Caloric Intake Prevents Metabolic Diseases in Mice Fed a High-Fat Diet (Hatori et al., 2012) — https://doi.org/10.1016/j.cmet.2012.04.019
Early Time-Restricted Feeding Improves Insulin Sensitivity, Blood Pressure, and Oxidative Stress Even without Weight Loss in Men with Prediabetes (Sutton et al., 2018) — https://doi.org/10.1016/j.cmet.2018.04.010
Early Time-Restricted Feeding Improves 24-Hour Glucose Levels and Affects Markers of the Circadian Clock, Aging, and Autophagy in Humans (Jamshed et al., 2019) — https://doi.org/10.3390/nu11061234
Calorie Restriction with or without Time-Restricted Eating in Weight Loss (Liu et al., 2022) — https://doi.org/10.1056/NEJMoa2114833

Watch the conversation: https://youtu.be/dKauIgX3IH4
Find us on YouTube: ⁠⁠⁠⁠https://youtube.com/levelshealth?sub_confirmation=1⁠⁠⁠

👋 Who we are:
Levels helps you understand your metabolic health with personalized data, expert guidance, and tools that connect your daily choices to measurable changes in your body. Our goal is to help you make better decisions about food, exercise, sleep, and long-term health.

Look for new shows every month on A Whole New Level, where we have in-depth conversations with thought leaders about metabolic health.

What if we could measure the human body the way engineers monitor a jet engine?
For nearly two decades, Stanford geneticist Dr. Michael Snyder has been trying exactly that. By continuously tracking everything he could possibly track in his body—genome sequencing, metabolites, wearable data, whole-body imaging—he has turned himself (and ~100 study subjects) into one of the world's largest experiments in continuous health measurement.
The point wasn’t mindless biohacking. It was to answer a question: Are there clinically useful signals in all that data?
This conversation suggests the answer is yes. Snyder’s experience and that of dozens of his study subjects offer some of the first real evidence that continuous measurement intelligently analyzed can reveal that you’re getting sick before you have symptoms, help find solutions to medical conditions that work for your unique biology, and help you refocus on prevention of disease rather than management.

Free course: Improve your metabolic health
Get our free email course on how glucose, nutrition, exercise, sleep, and measurement can help you build habits that support better energy and long-term health: ⁠⁠⁠⁠⁠⁠⁠⁠https://levels.link/wnl⁠

🎙️ About the Guest:
Dr. Michael Snyder is the Stanford W. Ascherman Professor of Genetics and director of the Stanford Center for Genomics and Personalized Medicine, where he previously chaired the Department of Genetics for 15 years.

📍What Dr. Michael Snyder & Mike Haney discussed:
03:10 Why Dr. Snyder started measuring healthy people instead of waiting for disease
09:18 Living in the future: tracking his own biology for nearly two decades
15:02 Why "normal" isn't enough: the importance of personal baselines
22:48 Discovering his own genetically unique form of type 2 diabetes
32:06 How wearables detected illness before symptoms appeared
39:18 Whole-body MRIs, incidental findings, and why change over time matters
46:55 Why everyone has abnormalities, but not everyone has disease
55:41 The promise of metabolomics, proteomics, and continuous biological sensing
1:06:52 Why AI is essential for interpreting thousands of health signals
1:15:34 What continuous measurement could mean for preventive medicine
1:23:10 The next decade of personalized health and what's still missing

🔗 Helpful Links:
Personal Omics Profiling Reveals Dynamic Molecular and Medical Phenotypes (Chen et al., Cell, 2012)
https://www.cell.com/cell/fulltext/S0092-8674(12)00965-3
Longitudinal Multi-omics Profiling Reveals Dynamic Personal Health Status (Price et al., Nature Biotechnology, 2017)
https://www.nature.com/articles/nbt.3695
The Integrative Personal Omics Profile (iPOP): A Decade of Progress in Precision Health (Snyder lab review)
https://www.nature.com/articles/s41576-021-00409-3
Wearable Sensor Data and Self-reported Symptoms for COVID-19 Detection (Mishra et al., Nature Biomedical Engineering, 2020)
https://www.nature.com/articles/s41551-020-00640-6
All of Us Research Program
https://allofus.nih.gov/
Connect with Dr Michael Snyder: https://med.stanford.edu/snyderlab.html

Watch the conversation: https://youtu.be/eUEYmLC1CcE
Find us on YouTube: ⁠⁠⁠https://youtube.com/levelshealth?sub_confirmation=1⁠⁠

👋 Who we are:
Levels helps you understand your metabolic health with personalized data, expert guidance, and tools that connect your daily choices to measurable changes in your body. Our goal is to help you make better decisions about food, exercise, sleep, and long-term health.

Look for new shows every month on A Whole New Level, where we have in-depth conversations with thought leaders about metabolic health.