The thread shows strong skepticism toward the AI recommendation due to lack of human trial data on several compounds, excessive injections, and unclear benefits. Participants advise simplifying to Retatrutide plus Tesamorelin, question the value of AOD and certain blends, and stress that AI needs heavy prompting with real studies to avoid hallucinations or overly agreeable answers. The original poster shares age, stats, and routine but receives advice to prioritize exercise and the two main compounds while dropping unproven items.
What this discussion establishesWhether LLMs can ever be trusted for peptide guidance even with good prompts, versus the view that they remain unreliable for anything medical
Still openWhether adding the third peptide or any of the lesser-tested compounds would actually improve results for this 56-year-old user beyond Retatrutide and Tesamorelin alone
Nothing here is advice.
Alternating those three because they hit overlapping metabolic paths. How would you improve this stack? Not claiming it is correct, just that I'm unsure and want input from people who know more.
If the AI said so then it has to be correct.
I already suspect it might be wrong, which is why I'm asking here.
That schedule means a lot of injections. You will go through needles fast. Sometimes fewer compounds work better anyway. You still have not said what result you are chasing.
Which model did you use? What outcome are you after with these peptides? I run ChatGPT in scholar mode with extra instructions for accuracy and tell it not to agree unless it can cite evidence. It would never suggest that combination and would instead point out there are no human trials showing safety or benefit for general health or most disease states, except for one specific use of tesamorelin. Asking separately about the evidence level for each one produces more cautious and accurate replies, though hallucinations still happen. Base models often just try to please the user rather than stay precise.
I once told an AI it was lying and it offered to dig deeper and maybe revise its answer.
The usual goals: drop fat, add muscle, sleep better, feel healthier overall.
More precise suggestions would need age, height, starting weight, target weight, current weight, how long the loss took, any other substances or health issues, and training details.
No human data supports AOD, get rid of it. The other one only raises appetite, which seems counterproductive. The low dose of the blend probably does nothing useful. Keep the reta.
Thanks for the detailed reply. Age 56, height 6'5", started at 301, aiming for 240, now at 277 after ten weeks. The peptides are the ones listed earlier. Training is lifting five or six days plus treadmill walking, running or hiking three or four days. Diet is mostly protein in a solid deficit. Health is good aside from carrying extra weight. Sleep is excellent.
The AI was just making things up.
Keep it straightforward. With the progress you already have and solid sleep, stay with reta and tesa first thing in the morning on an empty stomach. Once visceral fat is down, and only if no joint pain or fluid retention shows up, you could add the other one if you want. In the meantime try 30 g collagen plus vitamin C and hyaluronic acid each day, or look at glutamine and glucosamine with chondroitin.
Thirty-six injections in a week sounds painful, not for me.
It took effort to decode the abbreviations. I think I got most of them except one. AOD looks ineffective from the available research and may even be counterproductive by killing fat cells, which could worsen insulin resistance or leave uneven fat loss. The blend contains several interesting peptides but none have human testing and the mixture itself has never been studied even in animals. Taking an untested item that changes thousands of genes feels risky on principle. If it really did what is claimed for ligaments and skin it could do the same elsewhere with unknown effects.
They must be using a particular model known for weaker answers.
So drop tesamorelin? Appreciate the feedback.
Note that I am an AI and not a doctor. Consult a medical professional for personal advice. Here is what I received when I asked an AI about peptides.
You have to give it different instructions. It avoids giving doses on its own, but if you list the peptides and amounts it will build a schedule and explain the reasoning.
Large language models output probabilities, not facts. Stop using them for anything medical.
Probabilities are fine with me. I make decisions that way every day. Doctors do the same.
Doctors do not base decisions on letter and word probabilities.
What result are you aiming for with this stack? When using AI, first give it solid sources such as PubMed papers and instruct it to use only that material. Random queries to any model tend to be unreliable. I use NotebookLM and Zotero to control the input data.
AI can be helpful if you stay critical. Once it gave wrong information on recording laws and only corrected itself after ten minutes of pushback. Challenge it, ask for sources, and try multiple models. One told me not to run the stack because it was risky polypharmacy, another gave actual suggestions, and a third even pointed toward steroids.
The easiest way to get decent pharmacology answers is to use the research mode in ChatGPT. It usually avoids recommending untested compounds. More detailed prompts can improve accuracy further, and those prompts can be reused in new chats.
At least some of your education is still coming in handy.
That is exactly the approach I take.
Answer quality depends heavily on question quality. With good prompts it can reach specialist-doctor level on pharmacology. People without medical training often leave out key details or use imprecise language, which causes the model to make bigger mistakes such as missing urgent issues. When a doctor uses it the results are usually stronger because the input is better structured.
Any thoughts on the actual stack and ways to make it more efficient?
Ranked by importance: exercise and lifting first, then reta, then tesamorelin. I remain doubtful about most popular peptides because they lack human testing and therefore cannot be considered safe or proven to do what is claimed, especially the blends. DSIP I have not reviewed closely. Melatonin is well studied and generally safe below 4 mg. Tesamorelin alone is simpler and has more human data than combining several growth-related compounds. Using three peptides instead of one raises the chance of side effects or allergic reactions and makes it harder to identify the cause if problems appear. The appetite-stimulating one could also work against fat loss.
If you have tried it, this topic is still open. Sign in to reply — it takes a minute.