Someone with a six-pack tells you carbs are poison. Two days later, someone with equally impressive abs credits carbs for transforming their body. Both have hundreds of thousands of followers. Both sound authoritative. Both could be completely wrong.
This is the standard experience for anyone trying to navigate nutrition online — and it has nothing to do with gullibility. It's the result of predictable cognitive biases that the structure of social media is uniquely good at exploiting.
The confidence trap
Psychologists describe a well-documented confidence-competence gap: the people most certain about a topic are not always the most knowledgeable. Sometimes the relationship runs the other way. Researchers who have spent years in a field know how much remains unsettled. Someone who has read a handful of studies under good lighting in a home gym can project total certainty — because they genuinely don't know enough to doubt themselves.
This dynamic is particularly damaging in nutrition, a field with genuine, unresolved disagreements among credentialed scientists. When a TikTok creator strips out all that nuance and delivers a verdict with complete conviction, the brain interprets the confidence as credibility. They are not the same thing, but nothing in the viewing experience helps you separate them.
The likability illusion
The halo effect is well established in psychology: when someone is attractive, charismatic, funny, or relatable, the brain automatically upgrades its assessment of their competence and trustworthiness. This is not a conscious choice. It happens automatically.
Nutrition content creators who grow large audiences tend to be genuinely likable — they share vulnerability, they're entertaining, they feel like friends rather than broadcasters. That quality drives engagement. It has no relationship to whether their advice is accurate. You can be the most relatable person on the internet and still be completely wrong about intermittent fasting.
The complication is that some highly likable nutrition communicators are also credible. Some are not. The brain, busy liking someone, struggles to run that second check.
The credentials confusion
The natural response is to filter for credentials. The problem: credentials in nutrition are not uniform, and the terminology obscures the differences. A PhD in nutritional science and a certification from a weekend online course can both produce someone who calls themselves a "nutrition expert" and displays letters after their name.
Further, legitimate credentials don't immunize against bias. A registered dietitian sponsored by a supplement company may unconsciously recommend that supplement more than the evidence warrants — not out of dishonesty, but because humans are skilled at believing things that align with their interests. Confirmation bias operates on credentialed professionals the same way it operates on everyone else. A post from a qualified RD can still reflect undisclosed motivated reasoning, and nothing in the format signals that to the reader.
The sponsorship blindness problem
Research on media bias perception consistently shows that people are better at identifying bias in sources they distrust than in sources they already follow and like. When someone has earned your trust, sponsorship disclosures stop registering as warnings. The brain doesn't process "this person is being paid to say this" — it just processes the recommendation.
This is compounded by inconsistent disclosure practices. Sponsorships are sometimes buried in link descriptions, mentioned once casually, or framed in ways that don't activate the skepticism they should. The financial relationship between creator and brand quietly shapes what gets covered, what gets emphasized, and what gets framed positively — without the audience noticing the influence.
What actually works
Given that the brain is actively working against accurate evaluation here, the corrective requires deliberate slowdown.
Stop at confidence, not with it. Fast-scrolling mode produces snap judgments based on likability and certainty. When someone is making absolute claims about nutrition — "everyone should do this," "this is the truth" — that's a reason to pause, not a reason to trust. Real expertise in nutrition sounds like "the research suggests," "we don't fully understand this yet," and "results vary significantly between individuals." Hedged language is a credibility signal, not a weakness.
Verify credentials specifically. Registered Dietitian (RD or RDN) is a licensed credential requiring accredited education and ongoing professional standards. A PhD in nutrition science is a research credential with defined requirements. Many other titles in the space are self-assigned or backed by short certification programs with minimal standards. This information is usually findable if you look for it — it just requires looking.
Track the sponsorships actively. Check link pages and bios. Notice which brands appear repeatedly. Sponsorship doesn't mean dishonesty, but it does mean financial relationships are shaping content. The brain will try to minimize this information when you like the creator. Treat sponsorship patterns as data regardless.
Weight people who cite specific research. "Studies show" is nearly meaningless. A link to a specific study, a discussion of what that study actually measured, and an acknowledgment of where the evidence is limited — that's a different category of claim. Social media formats make this harder to do, which is one reason social media is a structurally poor environment for nutrition information.
Default toward nuance. Anyone claiming a single, universal answer to a nutrition question is almost certainly wrong. The field's actual experts are saying "it depends" frequently, because it usually does. Certainty at that scale is a sign of unfamiliarity with the complexity, not mastery of it.
The real problem
The human brain was not designed to evaluate nutrition claims from thousands of strangers with financial incentives to be persuasive. It was designed for fast decisions and social trust — tools that worked well in small communities and are badly calibrated for algorithmic feeds.
Across online nutrition spaces, the pattern repeats: people miss sponsorships in content they enjoy, extend credibility to confident voices without checking qualifications, and trust likable communicators past the point the evidence supports. This is not a failure of intelligence. It's a predictable output of cognition meeting a system optimized to exploit it.
Awareness of the biases helps but doesn't eliminate them. The practical floor is slowing down enough to run the checks your fast brain skips — credentials, sponsorships, certainty level, citation quality — and finding resources that have already done some of that vetting rather than starting from zero each time. (A searchable directory of nutrition influencers with credentials and sponsorships listed is available at https://nutrition-truth-68a3yth6h-alekos-projects-460515ef.vercel.app for exactly that purpose.)
The concrete rule: the more certain someone sounds about nutrition, the more carefully you should check what's actually behind that certainty.