Chris Williamson · 2026-08-15 · Chris Williamson (host), Daniel Lieberman
Harvard Professor: “I Tried Every Diet. This Is By Far The Worst.” - Daniel Lieberman
71 research-tied claims examined: 6 contradicted 7 overstated 5 context 43 supported 10 unverified
7 Overstated
Epidemiological studies indicate that one serving (about 85 grams) of meat per day increases cancer risk by approximately 13%, with a doubling of intake doubling the risk.
"there are studies which show that there's a dose-response effect between how much meat you eat and your risk of cancer. I think it's like one serving of meat, which I think they count as 85 grams, increases your your your risk of cancer over that over X years by I think 13%, and if you if you double that, you double the risk." (said at 0:07:15)
Epidemiological studies and dose-response meta-analyses have found that consuming one standard serving of red meat (typically defined as ~85 g or 100 g per day) is associated with an approximate 10% to 13% increase in relative risk for specific endpoints—such as total mortality (hazard ratio 1.13 per 85 g/day in large cohort studies) or colorectal cancer incidence (~12% increase per 100 g/day). However, the assertion that 'doubling the intake doubles the risk' conflates doubling the excess relative risk (an increase from ~13% to ~26% excess risk) with doubling the overall risk (a 200% or 2-fold risk). Furthermore, these risk elevations are observed for specific malignancies (notably colorectal cancer) and cause-specific mortality, rather than all cancers indiscriminately.
Rural populations in specific regions of Rwanda and Kenya consume at least 65% of their total dietary calories from carbohydrates and exhibit virtually no overweight or obesity.
"The area in Rwanda where I work, right, and the area in Kenya I work, people are getting at least 65% of their diet from carbs. At least 65%. That's a high-carb diet. And I dare you to find somebody with overweight or obesity in the area. They don't exist." (said at 0:49:08)
While rural diets in Rwanda and Kenya are predominantly carbohydrate-based (often deriving over 65% of total dietary energy from staple crops), the assertion that overweight or obesity 'don't exist' in these regions is overstated. Population-based surveys across East Africa (such as the WHO STEPwise and Demographic and Health Surveys) demonstrate that while rural rates of overweight and obesity are substantially lower than urban rates, overweight and obesity are documented and increasing in rural East African populations.
The National Institute on Aging (NIH) replicated the 20-year rhesus macaque calorie restriction study with a 30% reduction using standard monkey chow and found no significant difference in longevity.
"So the NIH redid the experiment and and this time they had them on the same monkey chow diet... And and one had a third, 30% less food than the other. And guess what? No difference, zilch in terms of health and longevity." (said at 0:55:13)
The claim bundles two assertions regarding the National Institute on Aging (NIA) long-term 30% calorie restriction (CR) study in rhesus macaques. The assertion that CR did not yield a statistically significant longevity or overall survival extension in the NIA cohort is supported by published findings (Mattison et al., 2012). However, claiming there was 'zilch in terms of health' overstates the lack of benefit: both the 2012 NIA report and a 2017 comprehensive joint analysis confirmed that CR resulted in clear health benefits, delaying age-related morbidity and improving metabolic parameters in monkeys.
Children who grow up in households with dogs have a lower incidence of autoimmune diseases and allergies.
"Have you seen the data around children that grow up in houses that have got dogs in? Yeah, exactly. And they're much less likely to get autoimmune diseases." (said at 1:27:16)
The speaker claims that children who grow up in households with dogs are "much less likely to get autoimmune diseases" and allergies. While there is evidence that early-life dog exposure is modestly associated with a reduced risk of certain atopic conditions like atopic dermatitis (eczema; RR ≈ 0.82), large pooled meta-analyses of birth cohorts (including over 77,000 mother-child pairs) demonstrate no statistically significant overall association between early-life dog ownership and childhood asthma or allergic sensitization (OR 0.92, 95% CI 0.85–1.01 and OR 0.93, 95% CI 0.57–1.54, respectively). Furthermore, broad claims that household dogs dramatically protect against autoimmune diseases as a whole overstate the current epidemiological literature, which shows modest, mixed, or disease-specific associations rather than a broad, substantial protective effect.
Approximately 25% to 30% of American teenagers are classified as obese.
"I think in the United States, something like 30% of American teenagers are now classified—or 25% are classified as being obese." (said at 1:32:18)
According to nationally representative surveillance data from the Centers for Disease Control and Prevention's National Health and Nutrition Examination Survey (NHANES), the prevalence of obesity (defined as BMI-for-age at or above the 95th percentile) among US adolescents aged 12 to 19 years is approximately 20% to 22%, rather than 25% to 30%. While specific demographic subgroups (such as certain racial, ethnic, or low-income groups) experience higher rates, the overall national prevalence of obesity among American teenagers remains below 25%. Figures in the 30% or higher range typically describe the combined prevalence of overweight and obesity rather than obesity alone.
Grinding peanuts into peanut butter increases available energy extraction by about 25% compared to consuming whole peanuts.
"And so, for example, if you take peanuts and you measure the content of the energy in your peanuts, and now you grind those peanuts up into peanut butter, which, by the way, is the world's greatest food, you get way more energy, about 25% more energy, than if you ate the peanuts whole" (said at 1:40:16)
The directional claim that grinding whole peanuts into peanut butter increases available (metabolizable) energy is correct: mechanical processing and roasting break open intact plant cell walls, releasing encapsulated lipids and increasing lipid bioaccessibility. A systematic review of human and in vitro studies confirmed that metabolizable energy is significantly influenced by physical form (flour > chopped > whole nuts) and processing (butter > roasted > raw). However, stating that peanut butter yields approximately 25% more available energy than whole peanuts overstates the magnitude of difference measured in metabolizable energy studies.
Hexane is a neurotoxin used as a solvent in oil extraction, and non-petroleum workers show detectable levels of hexane in their urine from consuming solvent-extracted soy products or seed oils.
"hexane is is a toxin. It's a neurotoxin. And um and and studies have shown that you can actually measure people's urine and people who who who are not in the petroleum industry and they have hexane in their urine and it's probably coming from soy or from non-expeller-pressed seed oil." (said at 1:44:44)
While n-hexane is an established neurotoxin used industrially in solvent extraction of edible oils, the claim that studies link detectable urinary hexane or its metabolites in non-industrial populations to the consumption of soy or seed oils is unsupported. In occupational settings, high inhalation exposure to n-hexane can cause peripheral polyneuropathy, and urinary 2,5-hexanedione is used as a biomarker for occupational monitoring. However, commercial refining and desolventization of edible oils reduce residual hexane to negligible, trace levels. In the general non-occupational population, background exposure to volatile organic compounds like hexane is predominantly driven by ambient air pollution, vehicle exhaust, and gasoline vapors, rather than dietary intake of seed oils or soy products.
Unverified means no publication matching the claim was located; it does not prove the claim false. Spotted an error? See the corrections policy - disputes from the people quoted are prioritized.