top of page
Search

The Shortcut Tax: What GLP-1s, AI Programs, and Peptides Actually Cost You

Writer: workinonmysavage
workinonmysavage
Oct 13
7 min read

You've felt the pull. A shot that melts the weight off. An app that writes your program in thirty seconds. A peptide your buddy swears healed his knee in a week. When you're juggling a career, a family, and a body that doesn't bounce back like it did at 25, a shortcut doesn't just look tempting. It looks smart.


So here's the question this piece answers: what are you actually trading when you take one?


Not "are these things fake." They're not. GLP-1 medications, AI-guided training, and recovery peptides each have something real behind them. But every one of them comes with a trade-off or an evidence gap, and if you're training for the long game, you need to know exactly where those sit before you sign up.


What people say defines the future of health


At the end of 2025, Life Time surveyed 750 consumers and members about where health and fitness are headed. When asked which trend will most define 2026, longevity came out on top at 37.8%. GLP-1 medications and peptides came second at 24.4%, and AI-guided training landed third at 14.6% [1].


One honest caveat: that question measured what people expect to trend, not what they'd personally choose. But it's still telling. Given the drug, the algorithm, and the long game, the long game won.


Let's look at why that instinct holds up.


GLP-1s: they work, and they take something with them


Let's give credit where it's due. These drugs produce real weight loss. In the STEP 1 trial, semaglutide drove roughly 15% average body-weight loss [2]. In SURMOUNT-1, tirzepatide reached about 21% at the highest dose [3]. For a lot of people, that's life-changing.


Here's the part the commercials skip. When researchers scanned body composition in those trials, lean mass made up about 40% of the weight lost on semaglutide and about 25% on tirzepatide [4].


Before you panic, two details matter:


  • Lean mass isn't the same as muscle. It also includes water, glycogen, and organ tissue, so the share that's actually skeletal muscle is smaller [4][5].

  • Dieting alone does this too. Weight lost through calorie restriction is typically 20–30% lean tissue [4][5].


So the drug isn't uniquely destroying muscle. It's doing what fast weight loss always does, just faster and at a bigger scale. For the aging athlete, that's the point: muscle is your currency for strength, metabolism, and independence later in life. Lose weight without protecting it, and you've paid a tax you'll feel for decades.


The medication can shrink you. It can't build you.


AI programming: a decent draft, not a coach


AI tools are getting good at spitting out a training plan. Researchers at the University of Würzburg had experienced coaches grade strength and hypertrophy programs written by popular AI models across 27 criteria. The verdict: moderate quality. The authors concluded that expert supervision is still needed, because the models can't account for how an individual responds, recovers, or gets hurt [6].


A second study from the same group found the plans got better when the prompts carried more detail [7]. Which makes sense: garbage in, garbage out.


That's the trade-off. AI can hand you a reasonable template. It can't see that your left hip has been cranky since Tuesday, that you slept four hours, or that your shoulder doesn't tolerate overhead pressing anymore. An algorithm can program you. It can't coach you.


Peptides: big buzz, thin human data


Now the one that's everywhere in combat sports right now: BPC-157, usually stacked with TB-500 in what people call the "Wolverine stack."


I'll be straight with you, because this is personal. This past Labor Day I hurt my right knee badly enough that I couldn't walk on it overnight (the full injury story is here), and the peptide recommendations started rolling in almost immediately. So I went to the research. Here's what's actually there.


BPC-157. A 2025 systematic review in HSS Journal pulled together everything published on BPC-157 in orthopaedic sports medicine. Of 36 studies, 35 were preclinical, meaning animals and cell cultures. One was in humans [8]. That single human study was a retrospective look at 16 people with knee pain, with no control group [9]. The review also found no clinical safety data in humans [8].


The animal results are genuinely interesting. But "promising in rats" isn't the same as "proven in you."


The regulatory picture. BPC-157 is not an FDA-approved drug. In July 2026, an FDA advisory committee voted 8–6 to recommend allowing compounding pharmacies to make it. That vote is non-binding, still needs FDA sign-off, and isn't a drug approval [10][11]. If you compete, there's a bigger issue: BPC-157 is banned at all times under WADA class S0 [11].


TB-500. The other half of the stack has even less behind it. A 2025 review in Arthroscopy found no human randomized trials for musculoskeletal use [12], and WADA prohibits it at all times [13].


So my decision came down to this: I don't know where the trade-off sits, and right now nobody else does either. That's my risk tolerance, not medical advice for yours.


What I did instead: the slower, evidence-grounded route


I went the long way: corrective-exercise rehab, staying on my feet, and a few targeted therapies. Imaging and an ultrasound exam ruled out fractures and showed my meniscus, ACL, PCL, and MCL were all intact. The diagnosis was a strain to the LCL.


Here's what the research says about that path:


  • Isolated grade 1–2 LCL injuries don't need surgery. In a study of 55 professional athletes treated without an operation, 100% returned to their pre-injury level, with a median recovery of 76 days [14]. That's the foundation: smart loading, staying mobile, and protecting the knee from the sideways forces that stress the ligament.

  • Laser (photobiomodulation). A meta-analysis found low-level laser therapy modestly reduced musculoskeletal pain compared with control [15]. Helpful for comfort, not a miracle.

  • Acoustic wave (shockwave). The strongest evidence is for plantar fasciitis and calcific tendinitis [16]. For fresh, acute injuries, the data is still limited [17].

  • Targeted cryo. A systematic review found little high-quality evidence that ice improves outcomes after acute soft-tissue injury [18]. It can feel good. It isn't what heals you.


Notice the pattern? The part with the strongest evidence is the least sexy one: keep moving, load it intelligently, and give it time. The gadgets are supporting cast.


I'm still mid-experiment. I'll know the real outcome when I'm back to full combat training. But I made the choice with my eyes open.


Myths worth retiring


"GLP-1s only burn fat." They don't. A meaningful share of the weight lost is lean tissue, which is why resistance training matters so much if you use them [4].


"An AI program is as good as a coach." Coaches who graded these plans called them moderate quality and said expert oversight is still needed [6].


"BPC-157 is proven to heal ligaments." In people, it isn't. Nearly all the evidence comes from animals, and the only human data is one small, uncontrolled study [8][9].


"The FDA approved BPC-157." It didn't. An advisory panel recommended allowing compounding, which is a different thing entirely [10][11].


"Fancy recovery tech is what gets you back." The best-supported piece of LCL recovery is conservative rehab and time. The devices help at the margins [14][15][17][18].


The Savage takeaway


Every shortcut has a price tag. Sometimes the price is lean mass. Sometimes it's a program that doesn't know your body. Sometimes it's a gamble on a compound with almost no human data. None of that makes these tools evil. It makes them tools, and tools don't replace the work.


So here's your challenge. Next time a shortcut is on the table, ask one question: what am I trading for this speed? If the answer is worth it to you, make the call with your eyes open. That's your body and your risk tolerance.


But know this: the needle can't build it, the algorithm can't build it. You build it. Stay savage.




This article is for educational purposes and isn't medical advice. Talk to a qualified healthcare provider before starting any medication, peptide, or rehab protocol.


Sources


  1. Life Time. 2026 Life Time Wellness Survey Results Are In (Dec 30, 2025). https://news.lifetime.life/2025-12-30-2026-Life-Time-Wellness-Survey-Results-Are-In-Strength-Training-and-Longevity-Lead-New-Year-Priorities-with-82-Focused-More-on-Wellbeing

  2. Wilding JPH et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). New England Journal of Medicine 2021;384:989–1002.

  3. Jastreboff AM et al. Tirzepatide once weekly for the treatment of obesity (SURMOUNT-1). New England Journal of Medicine 2022;387:205–216.

  4. Acibadem. Muscle loss on GLP-1 medicines: how much is lean mass. https://acibademinternational.com/blog/muscle-loss-on-glp-1-medicines-how-much-is-lean-mass-and-how-to-protect-it/

  5. Conte C, Hall KD, Klein S. Is weight loss–induced muscle mass loss clinically relevant? JAMA 2024;332(1):9–10.

  6. Havers T et al. A professional assessment of training plans for muscle hypertrophy and maximal strength developed by generative AI. Biology of Sport 2025;42(4):353–380. https://www.termedia.pl/A-professional-assessment-of-training-plans-for-muscle-hypertrophy-and-maximal-strength-developed-by-generative-artificial-intelligence,78,56339,0,1.html

  7. Havers T et al. Reproducibility and quality of hypertrophy-related training plans generated by GPT-4 and Google Gemini. Biology of Sport 2025;42(2):289–329. https://www.termedia.pl/Reproducibility-and-quality-of-hypertrophy-related-training-plans-generated-by-GPT-4-and-Google-Gemini-as-evaluated-by-coaching-experts,78,55336,0,1.html

  8. Vasireddi N et al. Emerging use of BPC-157 in orthopaedic sports medicine: a systematic review. HSS Journal 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12313605/

  9. Lee E, Padgett B. Intra-articular injection of BPC 157 for multiple types of knee pain. Alternative Therapies in Health and Medicine 2021;27(4):8–13.

  10. ABC News / Good Morning America. FDA advisory committee votes to add popular peptide BPC-157 to drug compounding list (Jul 23, 2026). https://www.goodmorningamerica.com/wellness/story/fda-advisory-committee-votes-add-popular-peptide-bpc-134913891

  11. Pillo. Is BPC-157 legal? (Oct 2, 2026). https://pillo.care/blog/is-bpc-157-legal

  12. DeFoor MT, Dekker TJ. Injectable therapeutic peptides in regenerative medicine and sports performance. Arthroscopy 2025, as summarized at https://superpower.com/guides/tb-500

  13. BSCG. TB-500: status, risks, and bans in sport and military. https://www.bscg.org/blogs/single/tb-500-status-risks-and-bans-in-sport-and-military

  14. Haslhofer DJ et al. Isolated clinically diagnosed grades 1–2 lateral collateral ligament injuries in elite athletes do not require surgery. Orthopaedic Journal of Sports Medicine 2025. https://www.repository.cam.ac.uk/items/9570084b-fe44-46ce-a209-07c935f2fbad

  15. Clijsen R et al. Effects of low-level laser therapy on pain in patients with musculoskeletal disorders. European Journal of Physical and Rehabilitation Medicine 2017;53(4):603–610. https://pubmed.ncbi.nlm.nih.gov/28145397/

  16. Speed C. A systematic review of shockwave therapies in soft tissue conditions. British Journal of Sports Medicine 2014;48(21):1538–1542. https://pubmed.ncbi.nlm.nih.gov/23918444/

  17. Zissler A et al. Extracorporeal shock wave therapy in acute injury care: a systematic review. Rehabilitation Process and Outcome 2018. https://journals.sagepub.com/doi/10.1177/1179572718765138

  18. Bleakley C, McDonough S, MacAuley D. The use of ice in the treatment of acute soft-tissue injury: a systematic review of randomized controlled trials. American Journal of Sports Medicine 2004;32(1):251–261.

 
 
 

Comments

Rated 0 out of 5 stars.
No ratings yet
Add a rating
bottom of page