The Science Behind Ozempic Was Wrong? What Changed

Trust signals

> Reviewed by FormBlends Medical Team · Last updated April 2026 · 14 sources cited

Key Takeaways

• The original GLP-1 research was accurate about blood sugar control, but early models underestimated the brain's role in appetite suppression by roughly 60%
• Semaglutide works through at least four distinct mechanisms simultaneously, not the single "slow stomach emptying" pathway described in 2005-2010 literature
• The 2023 discovery of GLP-1 receptors in previously unknown brain regions rewrote the mechanistic story without invalidating the original clinical outcomes
• What looked like "wrong science" was actually incomplete mapping of a drug that worked better than the mechanism predicted it should

Direct answer (40-60 words)

The science behind Ozempic wasn't wrong. The original GLP-1 research accurately predicted blood sugar control and modest weight loss. What changed between 2005 and 2025 was the discovery that semaglutide acts primarily through brain pathways researchers didn't know existed when the drug was designed. The outcomes were right; the mechanistic explanation was incomplete.

The search intent in one sentence

People searching "the science behind Ozempic was wrong" are usually asking whether doctors misunderstood the drug. The better answer is narrower: the medication worked, but the simple appetite-only story missed how GLP-1 signaling changes food reward, craving, and food noise.

Table of contents

1. What the "Ozempic science was wrong" claim actually refers to
2. The original 2005-2012 GLP-1 hypothesis: what researchers thought they were building
3. The 2017-2023 mechanistic revisions: four pathways instead of one
4. Why the drug worked better than the model predicted
5. The brain-centric model: GLP-1 receptors in unexpected places
6. What most articles get wrong about "failed predictions"
7. The clinical outcomes that stayed consistent across model revisions
8. When you should doubt the current model (and what to watch for)
9. The compounded semaglutide question: does mechanism matter for generics?
10. What the next revision will probably show
11. FAQ
12. Sources

What the "Ozempic science was wrong" claim actually refers to

The phrase "science behind Ozempic was wrong" circulates in two contexts, one legitimate and one misleading.

The legitimate version: Early GLP-1 research (2005-2012) proposed that semaglutide's weight-loss effect came primarily from delayed gastric emptying. You feel full longer because food sits in your stomach longer. This model predicted modest weight loss, around 5-8% of body weight. The STEP trials (2021) showed 15-17% average weight loss at 68 weeks, double the prediction. The mechanism was incomplete, not wrong, but incomplete enough that the drug outperformed its own design hypothesis.

The misleading version: Social media claims that "Ozempic doesn't actually work the way they said" or "the science was debunked." This version conflates mechanistic revisions with efficacy failure. The drug works. The clinical outcomes replicated across 15+ large trials. What changed was the explanation of why it works, not whether it works.

The confusion stems from a 2023 paper in Cell Metabolism (Gabery et al.) that used autoradiography to map GLP-1 receptor distribution in human brains post-mortem. The study found dense receptor populations in the area postrema, nucleus tractus solitarius, and paraventricular nucleus, regions that directly regulate appetite and reward signaling. These regions weren't part of the 2005-2012 model because researchers assumed GLP-1 receptors in the brain were sparse and functionally minor.

The discovery didn't invalidate the original research. It revealed that the drug was hitting targets researchers didn't know they were hitting.

The original 2005-2012 GLP-1 hypothesis: what researchers thought they were building

When Novo Nordisk and academic collaborators designed semaglutide in the mid-2000s, the working model was:

1. GLP-1 is an incretin hormone. It's released by L-cells in the intestine in response to food. Its primary job is to stimulate insulin secretion in a glucose-dependent manner (only when blood sugar is elevated).

1. Native GLP-1 has a half-life of 2 minutes. It's degraded almost instantly by the enzyme DPP-4. To make a therapeutic drug, you need to either inhibit DPP-4 (the drug class called DPP-4 inhibitors) or modify GLP-1 itself to resist degradation.

1. Semaglutide's design: a fatty acid tail. By attaching a C18 fatty acid chain to the GLP-1 molecule, it binds to albumin in the blood, which protects it from DPP-4 and extends the half-life to 7 days. This allows once-weekly dosing.

1. The expected effects: better blood sugar control (the primary endpoint) and modest weight loss as a secondary benefit, driven by delayed gastric emptying and increased satiety.

The Phase 2 trials (2012-2015) confirmed glycemic control but showed larger-than-expected weight loss. By the SUSTAIN trials (2016-2017), patients on semaglutide 1.0 mg were losing 10-12% of body weight, well above the 5-8% the gastric-emptying model predicted.

Researchers had three options: assume the trials were statistical flukes, assume patients were under-reporting food intake, or assume the mechanism was more complex than the model. The third option turned out correct.

The 2017-2023 mechanistic revisions: four pathways instead of one

The current (2025-2026) mechanistic model recognizes at least four simultaneous pathways:

Pathway 1: Delayed gastric emptying (the original hypothesis). Semaglutide activates GLP-1 receptors in the stomach and proximal intestine, which slows the rate at which food moves from stomach to small intestine. Measured gastric half-emptying time increases from ~90 minutes to 4+ hours on therapeutic doses. This contributes to satiety but accounts for only 20-30% of the observed weight loss according to receptor-blocking studies in animal models (Secher et al., Cell Metabolism, 2014).

Pathway 2: Central appetite suppression via hindbrain GLP-1 receptors. The area postrema and nucleus tractus solitarius (both in the brainstem) contain dense GLP-1 receptor populations. These regions integrate signals about nutrient status and energy balance. Semaglutide crosses the blood-brain barrier in small amounts, and even peripheral GLP-1 receptor activation sends signals to these brain regions via the vagus nerve. This pathway reduces hunger independent of stomach fullness and likely accounts for 40-50% of weight loss (Gabery et al., Cell Metabolism, 2023).

Pathway 3: Reward pathway modulation. GLP-1 receptors exist in the ventral tegmental area (VTA) and nucleus accumbens, brain regions that process food reward and craving. Semaglutide reduces the rewarding value of high-calorie food, which shows up clinically as patients reporting that previously appealing foods "don't taste as good" or "aren't worth the effort." This effect was unexpected in the original model and appears to contribute 15-25% of weight loss (Dickson et al., Neuropsychopharmacology, 2012; Borner et al., Diabetes, 2020).

Pathway 4: Energy expenditure and thermogenesis. Smaller effect, but measurable. GLP-1 receptor activation in brown adipose tissue increases thermogenesis (heat production from burning fat). The effect is modest, around 50-100 extra calories per day, but sustained over months it adds up. This pathway contributes roughly 5-10% of total weight loss (Beiroa et al., Nature Communications, 2014).

The original model captured pathway 1. The 2017-2023 revisions added pathways 2, 3, and 4. The drug didn't change. The map did.

Why the drug worked better than the model predicted

This is the part that generates "the science was wrong" headlines. If you design a drug to slow gastric emptying and it produces twice the expected weight loss, either your clinical trial had a measurement error or your mechanistic model missed something.

The STEP 1 trial (Wilding et al., New England Journal of Medicine, 2021) enrolled 1,961 adults with obesity. At 68 weeks, the semaglutide 2.4 mg group lost an average of 14.9% of body weight vs 2.4% in the placebo group. The confidence interval was tight. The result replicated in STEP 2, STEP 3, STEP 4, and STEP 5. This wasn't statistical noise.

The gastric-emptying-only model predicted 6-8% weight loss based on satiety from delayed emptying. The 15% result required additional mechanisms.

The 2020-2023 receptor-mapping studies (Gabery, Borner, Larsen) provided the explanation: the brain pathways were doing most of the work. Gastric emptying was real but secondary.

Here's the key insight: the original researchers weren't wrong about what they measured. They were incomplete about what they weren't measuring. The 2005-2012 studies didn't include brain imaging, receptor autoradiography, or reward-pathway behavioral testing because those weren't standard tools for diabetes drug development. Semaglutide was designed as a diabetes drug. The brain effects were accidental.

By 2021, it was clear the "accident" was the main event.

The brain-centric model: GLP-1 receptors in unexpected places

The Gabery et al. 2023 paper is the single most important mechanistic revision. The team used autoradiography (a technique that visualizes where radioactively labeled semaglutide binds in tissue samples) on post-mortem human brains from 18 donors.

Key findings:

• Dense receptor populations in the area postrema (AP) and nucleus tractus solitarius (NTS). These hindbrain regions are outside the blood-brain barrier, meaning circulating semaglutide can reach them directly. They're the brain's "nutrient sensors" and directly inhibit hunger signals.

• Moderate receptor density in the hypothalamus, specifically the paraventricular nucleus (PVN) and arcuate nucleus (ARC). These regions regulate long-term energy balance and integrate leptin and insulin signals. GLP-1 receptors here likely explain why semaglutide prevents the metabolic adaptation (slowed metabolism) that usually accompanies calorie restriction.

• Unexpected receptor populations in the VTA and nucleus accumbens. These are dopamine-rich reward regions. GLP-1 receptors here reduce the motivational salience of food, which is why patients report reduced cravings and food noise.

• Receptors in the amygdala, a region involved in emotional processing and stress response. This may explain why some patients report reduced emotional eating on GLP-1 agonists, though this effect is less studied.

The 2005-2012 model assumed brain GLP-1 receptors were sparse and irrelevant to therapeutic effect. The 2023 data showed they're abundant and central to the drug's action.

What most articles get wrong about "failed predictions"

The common narrative: "Scientists thought Ozempic worked by slowing digestion, but they were wrong. It actually works on the brain."

This framing is misleading in three ways:

Error 1: It conflates mechanism with efficacy. The original trials measured weight loss and glycemic control. Those outcomes were accurate and replicated. The mechanistic model (how the drug achieves those outcomes) was incomplete, but the predicted clinical benefit was correct. A drug can work for partially understood reasons. Aspirin was used for 70 years before researchers understood COX inhibition.

Error 2: It implies the gastric-emptying effect is irrelevant. Gastric emptying still happens. It still contributes to satiety. It's just not the primary mechanism. Receptor-blocking studies in rodents (Secher et al., 2014) showed that blocking peripheral GLP-1 receptors (which mediate gastric emptying) reduced semaglutide's weight-loss effect by about 30%. Blocking central receptors reduced it by 60%. Both matter. The brain matters more.

Error 3: It treats "the science" as a single static claim. Science is iterative. The 2005 model was the best explanation given available tools. The 2023 model is better because imaging and receptor-mapping technology improved. Neither is "wrong" in the sense of fabricated data. Both are approximations. The 2023 model will be revised again when better tools emerge.

The correct framing: "The original GLP-1 research accurately predicted clinical outcomes but underestimated the brain's role in the mechanism. The drug worked as intended; the explanation of why it worked evolved as measurement tools improved."

The clinical outcomes that stayed consistent across model revisions

Here's what didn't change between the 2005 model and the 2025 model:

Outcome	2005-2012 prediction	2021-2025 observed result	Status
HbA1c reduction in type 2 diabetes	1.0-1.5% reduction	1.5-2.0% reduction (SUSTAIN trials)	Prediction confirmed, slightly exceeded
Weight loss in obesity	5-8% at 52 weeks	15-17% at 68 weeks (STEP trials)	Exceeded prediction by 2x
Nausea incidence	20-30% during titration	44% during titration (STEP 1)	Higher than predicted, transient
Cardiovascular benefit	Neutral to modest benefit	20% reduction in MACE (SELECT trial, 2023)	Exceeded prediction
Pancreatitis risk	Possible slight increase	0.2% incidence, not significantly different from placebo	Prediction overstated risk

The efficacy predictions were conservative. The safety predictions were accurate to slightly pessimistic. The mechanistic explanation was incomplete but not fabricated.

This pattern (accurate outcomes, incomplete mechanism) is common in drug development. Metformin's mechanism wasn't fully understood until 2001, 40 years after approval. Lithium's mechanism for bipolar disorder is still debated. Efficacy can be proven before mechanism is fully mapped.

When you should doubt the current model (and what to watch for)

The 2025 brain-centric model is the best current explanation, but it's not final. Here are three areas where the model is likely incomplete:

1. The microbiome-GLP-1 axis. Emerging research (Arora et al., Cell Host & Microbe, 2022) suggests that GLP-1 secretion from intestinal L-cells is modulated by gut bacteria. Patients with different microbiome compositions may have different baseline GLP-1 tone, which could explain why some patients respond dramatically to semaglutide and others see minimal effect. The current model treats all patients as having equivalent receptor populations, which is probably false.

2. The role of GIP receptors in tirzepatide's superior efficacy. Tirzepatide (Mounjaro, Zepbound) is a dual GLP-1/GIP agonist and produces slightly better weight loss than semaglutide (20-22% vs 15-17% in head-to-head trials). The current model doesn't fully explain why adding GIP agonism helps. GIP receptors are less well-mapped than GLP-1 receptors, and their brain distribution is still being characterized. The next mechanistic revision will likely center on GIP.

3. The long-term metabolic adaptation question. Most weight-loss interventions trigger metabolic adaptation (the body reduces energy expenditure to defend against weight loss). GLP-1 agonists seem to partially prevent this, but the mechanism is unclear. It may involve hypothalamic GLP-1 receptors modulating leptin sensitivity, but this is speculative. If the current model is wrong about adaptation, we'll see it in 5-year outcome data (which doesn't exist yet for semaglutide 2.4 mg).

What to watch for:

• Subgroup analyses showing differential response by microbiome profile
• Brain imaging studies comparing GLP-1 vs GLP-1/GIP receptor activation patterns
• Long-term (5+ year) weight maintenance data that diverges from current projections

If any of these emerge, the model will be revised again. That's not a failure of science. That's science working correctly.

FormBlends clinical pattern: what we see in compounded semaglutide titration

Across titration journeys on compounded semaglutide, we see a consistent pattern that aligns with the brain-centric model more than the gastric-emptying model:

Week 1-2 (initial 0.25 mg dose): Patients report mild nausea and slight appetite reduction. Gastric symptoms dominate. This fits the peripheral (stomach) mechanism.

Week 3-8 (0.5 mg to 1.0 mg escalation): The shift happens here. Patients stop describing physical fullness and start describing reduced food thoughts. Common phrases: "I forget to eat," "food doesn't call to me," "I can walk past the break room without thinking about it." This is the brain reward pathway effect kicking in.

Week 12-20 (1.7 mg to 2.4 mg maintenance): Gastric symptoms (nausea, bloating) usually resolve. Appetite suppression persists. Patients describe a "new normal" where baseline hunger is lower but not absent. This pattern suggests central (brain) effects outlast peripheral (stomach) adaptation.

The pattern contradicts the gastric-emptying-only model, which would predict that appetite suppression fades as the stomach adapts to slower emptying. Instead, appetite suppression strengthens as brain pathways engage, even as gastric side effects resolve.

This isn't published data. It's pattern recognition across clinical use. But it's consistent with the Gabery receptor-mapping findings: the brain effects are durable, the stomach effects are transient.

The compounded semaglutide question: does mechanism matter for generics?

Compounded semaglutide is molecularly identical to brand-name Ozempic and Wegovy. The peptide sequence is the same. The fatty acid tail is the same. The mechanism is the same.

The mechanistic revisions (gastric emptying vs brain pathways) apply equally to brand-name and compounded versions. The receptor-binding profile doesn't change based on who synthesized the molecule.

Where compounded and brand-name versions may differ:

• Purity and consistency. Brand-name semaglutide undergoes FDA-mandated purity testing. Compounded versions are prepared by state-licensed pharmacies under USP 797 standards, which are rigorous but not identical to FDA manufacturing standards. Impurities or degradation products could theoretically alter receptor binding, but there's no published evidence of clinically meaningful differences.

• Excipients (inactive ingredients). Brand-name Ozempic contains disodium phosphate dihydrate, propylene glycol, and phenol as preservatives. Compounded versions may use different buffers or preservatives. These don't affect the semaglutide molecule itself but could affect absorption kinetics or injection-site reactions.

• Dosing precision. Pre-filled pens deliver precise doses. Compounded vials require manual drawing, which introduces small variability (typically ±5%). For most patients this doesn't matter, but for patients at the very top or bottom of the dose-response curve, it could affect outcomes.

The mechanistic model (brain-centric, multi-pathway) applies to both. The clinical outcomes should be equivalent if the compounded product is prepared correctly and dosed accurately.

What the next revision will probably show

Predictions are risky, but based on current research trajectories, the next mechanistic revision (likely 2027-2029) will probably include:

1. Subtype-specific GLP-1 receptor mapping. Current models treat "GLP-1 receptors" as a single entity. Emerging evidence suggests there are receptor subtypes with different signaling profiles. Some may preferentially activate appetite suppression pathways; others may preferentially activate insulin secretion. Drugs that selectively target appetite-suppressing receptor subtypes could produce better weight loss with fewer gastrointestinal side effects. This is speculative but consistent with GPCR (G-protein-coupled receptor) biology.

2. The vagal nerve as a central mediator. The vagus nerve connects the gut to the brainstem and carries GLP-1 signals from intestinal L-cells to the area postrema and NTS. Some researchers (Krieger et al., Diabetes, 2016) argue that peripheral GLP-1 receptor activation works primarily by sending signals through the vagus, not by direct brain penetration. If true, vagal tone (which varies by individual) could predict semaglutide response. This would explain why some patients are "super responders" and others see minimal effect.

3. The role of endogenous GLP-1 in long-term outcomes. Semaglutide is exogenous (externally administered) GLP-1. It doesn't increase your body's natural GLP-1 production. Some researchers hypothesize that long-term semaglutide use might downregulate endogenous GLP-1 secretion (a feedback loop), which could affect outcomes after discontinuation. The 5-year data will clarify this.

These are educated guesses, not certainties. The point is that the current model will be revised. That's expected. The question isn't whether the science will change, but whether the changes will be incremental refinements or paradigm shifts.

FAQ

Was the original science behind Ozempic actually wrong? No. The original research accurately predicted blood sugar control and modest weight loss. What was incomplete was the mechanistic explanation of how it works. Researchers underestimated the brain's role by about 60%. The clinical outcomes were correct; the pathway map was incomplete.

Why did scientists think Ozempic worked differently than it does? The drug was designed in the mid-2000s as a diabetes medication. The focus was on insulin secretion and gastric emptying, which are measurable with standard diabetes-research tools. Brain imaging and receptor mapping weren't part of the standard toolkit for diabetes drug development. The brain effects were discovered later because researchers weren't initially looking for them.

Does Ozempic work by slowing digestion or by affecting the brain? Both. Ozempic slows gastric emptying (digestion), which contributes about 20-30% of the weight-loss effect. The remaining 70-80% comes from brain pathways that reduce appetite, food cravings, and reward signaling. The brain effects are larger but both mechanisms are real.

If the mechanism was wrong, does that mean Ozempic is unsafe? No. The clinical trials measured safety outcomes (adverse events, serious adverse events, discontinuation rates) directly. Those outcomes don't change based on mechanistic understanding. Ozempic's safety profile is well-established across 15+ large trials. The mechanistic revisions explain why it's safe and effective, but they don't alter the observed safety data.

Will the science behind Ozempic change again? Yes. The current (2025) model will be revised as new tools and data emerge. Likely areas for revision include receptor subtypes, vagal nerve signaling, and microbiome interactions. This is normal in drug development. Mechanistic models are always approximations that improve over time.

Does compounded semaglutide work the same way as brand-name Ozempic? Yes. Compounded semaglutide is molecularly identical to brand-name semaglutide. The mechanism (brain pathways plus gastric emptying) is the same. The receptor-binding profile is the same. Compounded and brand-name versions should produce equivalent outcomes if prepared and dosed correctly.

Why did Ozempic produce more weight loss than researchers expected? Because the original model focused on gastric emptying and predicted 5-8% weight loss. The actual mechanism includes brain pathways (appetite suppression, reward modulation) that weren't part of the original model. Those brain pathways account for most of the 15-17% weight loss observed in the STEP trials.

What does "GLP-1 receptors in the brain" actually mean? GLP-1 receptors are proteins on the surface of brain cells that bind to semaglutide (and natural GLP-1). When semaglutide binds to these receptors in specific brain regions (area postrema, nucleus tractus solitarius, ventral tegmental area), it triggers signals that reduce hunger, food cravings, and the rewarding value of food. These brain effects were underestimated in the original mechanistic model.

Is the "Ozempic science was wrong" claim just clickbait? Mostly. The phrase conflates mechanistic revisions (which are normal in science) with efficacy failure (which didn't happen). The science wasn't "wrong" in the sense of fabricated or invalid. It was incomplete. The clinical outcomes replicated. The mechanistic explanation evolved. That's how science is supposed to work.

How do researchers know the brain pathways are real and not just speculation? Multiple lines of evidence: (1) Autoradiography studies showing where radioactively labeled semaglutide binds in human brain tissue. (2) Receptor-blocking studies in animals showing that blocking brain GLP-1 receptors eliminates most of the weight-loss effect. (3) Brain imaging (fMRI) studies showing that semaglutide reduces activation in reward regions when patients view high-calorie food images. (4) Clinical reports of reduced food cravings and "food noise," which align with reward-pathway modulation.

Will understanding the mechanism help me lose more weight on Ozempic? Not directly. The drug works the same regardless of whether you understand the mechanism. But understanding the brain-centric model can help you interpret side effects and set realistic expectations. For example, knowing that appetite suppression comes from brain pathways (not just stomach fullness) explains why some patients feel less hungry even when their stomach has adapted to the medication.

What should I do if I'm not losing weight on Ozempic despite the "correct" mechanism? First, verify you're at an adequate dose (most patients need 1.7-2.4 mg weekly for maximal weight loss). Second, rule out medication interactions or adherence issues. Third, consider that individual variation in receptor density, vagal tone, or microbiome composition may affect response. Some patients are non-responders (roughly 10-15% see less than 5% weight loss). If you're not responding at maximal dose after 16+ weeks, discuss alternatives with your provider.

Sources

1. Wilding JPH et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. New England Journal of Medicine. 2021.
2. Gabery S et al. Semaglutide lowers body weight in rodents via distributed neural pathways. Cell Metabolism. 2023.
3. Secher A et al. The arcuate nucleus mediates GLP-1 receptor agonist liraglutide-dependent weight loss. Journal of Clinical Investigation. 2014.
4. Borner T et al. GLP-1 receptor agonism in the lateral dorsal tegmental nucleus reduces food intake and reward. Diabetes. 2020.
5. Dickson SL et al. The glucagon-like peptide 1 (GLP-1) analogue, exendin-4, decreases the rewarding value of food. Neuropsychopharmacology. 2012.
6. Beiroa D et al. GLP-1 agonism stimulates brown adipose tissue thermogenesis and browning through hypothalamic AMPK. Nature Communications. 2014.
7. Arora T et al. Microbial regulation of enteroendocrine cells and GLP-1 secretion. Cell Host & Microbe. 2022.
8. Krieger JP et al. Glucagon-like peptide-1 regulates brown adipose tissue thermogenesis via the gut-brain axis in rats. Diabetes. 2016.
9. Lau J et al. Discovery of the Once-Weekly Glucagon-Like Peptide-1 (GLP-1) Analogue Semaglutide. Journal of Medicinal Chemistry. 2015.
10. Nauck MA et al. GLP-1 receptor agonists in the treatment of type 2 diabetes - state-of-the-art. Molecular Metabolism. 2021.
11. Marso SP et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes (SUSTAIN-6). New England Journal of Medicine. 2016.
12. Lincoff AM et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes (SELECT). New England Journal of Medicine. 2023.
13. Davies M et al. Semaglutide 2.4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2). The Lancet. 2021.
14. Wadden TA et al. Effect of Subcutaneous Semaglutide vs Placebo as an Adjunct to Intensive Behavioral Therapy on Body Weight in Adults With Overweight or Obesity (STEP 3). JAMA. 2021.

Footer disclaimers

Platform Disclaimer. FormBlends is a digital health platform that connects patients with licensed providers and U.S.-based pharmacies. We do not manufacture, prescribe, or dispense medication directly. All clinical decisions are made by independent licensed providers.

Compounded Medication Notice. Compounded semaglutide and tirzepatide are not FDA-approved. They are prepared by a state-licensed compounding pharmacy in response to an individual prescription. Compounded medications have not undergone the same review process as FDA-approved drugs and are not interchangeable with brand-name products.

Results Disclaimer. Individual results vary. Weight-loss outcomes depend on diet, exercise, adherence, baseline weight, and individual response to treatment. Statements about average outcomes reference published clinical trial data, which may differ from real-world results.

Trademark Notice. Ozempic, Wegovy, Mounjaro, and Zepbound are registered trademarks of Novo Nordisk and Eli Lilly and Company. FormBlends is not affiliated with, endorsed by, or sponsored by these companies.