Stanford's AI Finds a Body-Made Peptide That Curbs Appetite in Animals

The Pitch: Cheaper and More Targeted

Ozempic can cost more than $1,000 a month without insurance, so a naturally occurring compound that could seed cheaper or more targeted drugs deserves attention. Stanford researchers say AI helped them find one. The molecule is a 12-amino-acid peptide called BRP, and early animal results suggest it works differently from semaglutide in ways that matter.

Here's the thing: the price argument is what makes this story land with general readers, even though the science is the real headline. Semaglutide's visibility has put pressure on costs and on anyone trying to find an alternative. Similar drug discovery efforts typically face a long road from a promising lab result to something a patient can actually fill, so a single finding is a starting point, not a price tag.

Where the Study Sits

The work, published in March 2025, focused on peptide hormones, the signaling molecules that help the body regulate energy, hunger, and blood sugar. Much of the recent attention in this area has gone to drugs built on hormones the body already uses, so a naturally occurring candidate is a different kind of bet. The team programmed AI to map peptide fragments, and BRP stood out because it appeared to act like semaglutide, but in a narrower way. Here's what I'd watch for: "narrower" is the word doing the heavy lifting in that sentence.

Why AI Made the Difference

Traditional protein isolation struggles to separate peptide hormones, which are relatively rare, from the far more abundant fragments left behind when proteins break down. That's a search problem, and it's where the algorithm earned its keep. It scanned 20,000 protein-coding genes and narrowed the field to 373 prohormones containing 2,683 peptides. The team then zeroed in on 100 sequences likely to be bioactive in the brain, similar to GLP-1 drugs.

This is the part that usually gets lost in AI coverage. The algorithm didn't create anything new; it sorted sequences the body already produces (and this is the part most people overlook) and flagged the ones worth testing. Screening efforts like this generally trade breadth for confidence, which means the shortlist is only as good as the lab work that follows. If you're working in computational biology, the methodology is the section to read closely, because the pipeline matters more than any single hit.

What the Animal Data Shows

The researchers tested BRP in lean mice and miniature pigs, the latter of which more closely mirror human eating and digestion. An intramuscular injection given before feeding reduced food intake over the next hour by up to 50% in both models. Eating less over repeated treatments would be expected to reduce body fat, though that's an inference rather than a result these studies measured. Follow-up work on physiological and brain activity suggested BRP engages metabolic and neuronal pathways that differ from those triggered by GLP-1 drugs or semaglutide.

Lean mice are a common starting point, but they tell you little about appetite in people who are closer to the populations these drugs are meant for. Pigs are a better proxy for digestion, and that's why their inclusion here is more than a checkbox. Still, appetite results in animals have a mixed track record, so I'd treat a one-hour drop in eating as a signal rather than proof.

The real distinction between BRP and semaglutide comes down to where each one acts. Semaglutide's receptors show up in the brain, gut, and pancreas, which helps explain its broad effects, as Svensson noted. BRP's apparent focus on the hypothalamus, which controls appetite and metabolism, points toward a narrower target. Narrower can mean fewer side effects, but it can also mean less weight loss, and the animal data can't settle that trade-off.

What Comes Next

Svensson co-founded a company to pursue clinical trials, which is a common step when academic work starts moving toward a drug. Before BRP reaches people, the team is working on making its effects last longer, which could allow a more convenient dosing schedule, if the peptide proves effective at regulating human body weight. Human results haven't been published, and no release date has been confirmed. A company formed to pursue trials signals intent, but it isn't evidence of progress.

"The lack of effective drugs to treat obesity in humans has been a problem for decades," Svensson said. "Nothing we've tested before has compared to semaglutide's ability to decrease appetite and body weight. We are very eager to learn if (BRP) is safe and effective in humans." That frustration is hard to argue with, especially since, by her account, nothing her team had tested matched semaglutide. And yet the bar she's describing is a high one, and animal studies rarely clear it on their own.

Safety is the question that matters most here. A compound that works in mice and pigs can still cause problems in people, and animal dosing rarely captures long-term effects. Trials will need to show not just changes in appetite or weight but whether the narrower mechanism actually translates into a cleaner side effect profile. That's the implied advantage, and it's still unproven.

What This Means for the Field

If you're watching the obesity drug space, the practical takeaway is that discovery methods are becoming a story in their own right. Similar AI-driven screens could surface other peptides, and the bottleneck will likely shift from finding candidates to testing them. That's a healthier problem to have, but it's still a bottleneck, and it's expensive.

I'd be skeptical of anyone calling BRP a successor to semaglutide this early. It's an interesting lead with a plausible mechanism, and the animal data are suggestive. A one-hour reduction in eating in mice and pigs, though, doesn't tell us how a human body will respond over months of treatment. That gap is where most promising candidates have historically fallen apart.

What to watch: human safety data, whether the effects hold with repeated dosing, and whether the company's trial plans move forward on a public timeline. Natural compounds could also offer a cheaper route than fully synthetic drugs, but that's a hope until trials show it. The human data will decide whether BRP becomes a real successor to semaglutide or a useful lead that never leaves the lab.