In my previous post I lifted up an interesting ranked list of compounds with longevity potential that the researcher Nir Barzilai presented at the ARDD 2024 conference. Now I would like to share another very interesting overview which he presented at the conference around expanded indications on the top compounds in his ranked list. Important note is that the overview is still under progress and maybe later this year Nir will publish a paper on it. I have also in this overview that I present added rapamycin to it and some other expanded indications. The ones that have been added to the overview I have marked with the text ā€œ(new)ā€.

The thing I really like with this overview is that it tries to summarize things in a quite elegant way both the good and not so good parts. So really great work here, Nir! It would be very interesting to expand this overview and include the most common FDA approved drugs with longevity potential and maybe also basic lifestyle interventions and promising supplements. I think this type of overview helps to pinpoint where there are missing pieces and research studies need to be set up so that we as a field step by step get a full picture of this puzzle. Regarding Rapamycin, have I missed something in that column or a row? All feedback is welcome as always. Even the tough one.

8 Likes

To me what really stands out here is that Rapamycin is either very non-specific in its effects, or we just donā€™t know if there is an effect.

5 Likes

What strikes me, is that rapamcin really only has one (+), and thatā€™s flu/Covid prevention, whereas all the other drugs in this table have multiple (++) including the dodgy bisphosphonates. So rapamycin looks the weakest here, barely scoring a single win. Interesting.

Meanwhile of all the drugs in this table, it is only rapamycin which shows far and away the best lifespan extension in every species where it was tested. Next come SGLT2/1i, and thatā€™s about it, The best scoring drug in this table, by far is metformin, racking up (++) like there is no tomorrow - meanwhile it seems fairly inert in the lifespan extension race, by itself, and only notches a small win when combined with, yes, rapamycin.

So, what do we make of this? ā€œSecondary preventionā€ implies the prevention of premature mortality, not necessarily extension of max lifespan, so maybe thatā€™s whatā€™s shown here. Maybe metformin is the beeā€™s knees in this regard, at least according to this table. Still not sure how bisphosphonates fit in here, but itā€™s WIP (Work In Progress), I suppose.

Iā€™m very happy to learn of Kristerā€™s evident enthusiasm for this table, as I personally am unable to muster much enthusiasmā€¦ possibly because Iā€™m always most impressed by outcome studies, less by mechanistic speculation. As far as that goes, rapa is still king of the hill (in animal models), regardless of what this table claims. YMMV.

3 Likes

To me, the conclusion is that Rapamycin simply isnā€™t a very well supported intervention. Thereā€™s just WAY less of the type of data we need, compared to GLP-1A and SGLT2i drugs which have had massive clinical trials and now have tons of data across multiple disease models. Thatā€™s because thereā€™s commercial interest, profit to be made, and the drug companies would love to expand their indications. Itā€™s also trendy and sexy so a lot of research effort goes into it.

Rapamycin has a much less sexy history, was labelled as immunosuppressant and is tarnished by being associated with organ transplants. Studies are mostly limited to a few mouse and worm studies, some poorly-performed dog and monkey studies (some of which returned negative results). There are a few studies in humans, but theyā€™re often very limited. For example, even the famous Mannick study which established the whole ā€œonce weeklyā€ dosing paradigm, really isnā€™t a great or super conclusive study. Firstly, it wasnā€™t strictly Rapamycin that was tested. Secondly, the results show a somewhat inconsistent, unexplainable increase in immune response to some strains of a flu vaccine, which was different depending on dosing schedule. To me, it makes little sense that 5mg/week increases response to H1N1 and H3N2, but 20mg/week only works for H3N2. And even then, weā€™re talking about a 20% increase, which is nothing crazy.

However, the appeal is that Rapamycin seems to tap into something more fundamental about aging than the others. GLP-1A and SGLT2i are likely getting the most bang for buck by controlling glucose. Are they true longevity drugs? Maybe not IMO. As you said, maybe itā€™s secondary prevention by fixing problems and making people ā€œhealthierā€ across multiple systems. That, in turn, lowers risks of other diseases like cancer, simply by reducing some risk factors.

Unfortunately, what we really need is some benevolent multi-millionaire to fund a study of Rapamycin out of pure charity. Until then, I donā€™t see the situation improving much. Matt K obviously has the dog study, which will be cool, but itā€™s still not ideally powered or as big as we might like. Thereā€™s still an inherent huge amount of randomness and noise in biology. If we look at the ITP data, even though median and maximum lifespan were extended, some individual mice getting Rapamycin died before mice in the placebo group at all 3 sites. Nothing is guaranteed unfortunately.

Still, @Krister_Kauppi thanks for sharing!

8 Likes

I mostly agree. Though I think the animal evidence for rapa longevity is a bit better than your take, seems to me. In mice itā€™s pretty solid. In dogs/cats, obviously the longevity data simply isnā€™t there. Marmosets - once the study comes out officially, Iā€™ll make up my mind about those.

My considerations are around safety. The PEARL trial, among others seems to say rapa is pretty safe. As always, dosage is key. You might derive ā€œmodestā€ benefits from a modest dose. A modest dose is a safe dose. A high dose is like a race car - great speed/benefits, but much greater risk, including that for some, gunning the engine that aggressively will flip the whole car, and have some mice die prematurely compared to a placebo. So possible greater benefits at the cost of safety. Matt Kaeberlein pointed out something important: you might slow down the aging and die prematurely. Rapamycin may genuinely slow down the aging substantially, but also weaken some key systems (like immunity), which will kill you opportunistically. Thatā€™s something to ponder. And thatā€™s where dosage - benefits - risks comes in. So combine this with side-effect combatting drugs addressing possible lipid and glucose elevations (interestingly, seems might be even beneficial per the ITP), keep up with all the vaccinations and not go wild with unnecessary pathogen exposures.

The ideal is to gun the engine just enough to extract the maximum benefit in slowing aging, but not push it to the point of not making a corner (in our race car analogy). Trouble is, we donā€™t know where that dosage is, and for that matter might be individual for each of us. So a safer course of action might be to simply get in the car, and go somewhat faster but also not risk it almost at all, so middle of the track, no sharp corners. Maybe 30mg/week might be a huge benefit - with attendant risk. And maybe 3-6-9-12mg/week might be a modest benefit, with minimal risk. We donā€™t know. Itā€™s all a gamble. Up to you and your temperament. Are you a risk taker? Most of us taking rapa, are to some degree risk takers.

My approach is reasoned polypharmacy. Itā€™s still a gamble. I have garbage lipids. So I take a statin - pitavastatin (and might take more: bempedoic acid, ezetemibe etc.) - just to cover my premature MACE/CVD mortality/morbidity risk. I have trash glucose control (prediabetic), so I take empagliflozin, and might take more (pioglitazone, GLP-1RA etc.) again, to cover premature morbidity/mortality. Iā€™m looking to some telmisartan to possibly slightly lower my BP down the road. All these are really there to normalize my lifespan, not extend it, although oddly it seems such drugs (some statins, some SGLT2/1i, some BP meds) might actually see a slight longevity bump in animal models (mice). But Iā€™m not counting on that. And then, for longevity, slowing down of aging, Iā€™m throwing the dice with rapamycin at a modest dose. Itā€™s like a retiree, who has budgeted out his living expenses fully for a comfortable retirement, and then puts aside a small monthly stack to take to Vegas and spin for some luck. As long as the gambling money is strictly budgeted out, itā€™s harmless - you probably will lose it all, but have a small chance at winning a small fortune. No high stakes - no high doses - just safe, small stakes small payouts, for entertainment. That is why I focus on safety. And seemingly rapa is reasonably safe - PEARL tells us that at least 3mg/week seems safe, and actually shows some (very, very, very modest) benefits for women. If we take body size into account, my current 6mg/week seems safe too - but of course subject to careful monitoring and biomarker testing and tracking. Perhaps 20mg/week would give me bigger benefits (everolimus 20mg/week in Mannickā€™s study), but itā€™s already a bit too rich for my blood - in this department, Iā€™m strictly a small stakes recreational gambler. So Iā€™m sticking to 6mg - if feeling frisky, I might, might escalate to 10-12mg/week, based on companion animal (cats and dogs) data of 0.15mg/kg dosages, but then a hard stop - pending of course more data, if it comes out that a different dose is superior, then Iā€™ll adjust.

Polypharmacy means you have to research any possible DDI (Drug Drug Interactions), which Iā€™ve done for my stuff (pitavastatin, empagliflozin, telmisartan), and carefully monitor biomarkers.

But thatā€™s the approach - bread and butter of the proven drugs in extensive human trials and clinical experience, with the dessert topping of rapa in modest amounts. Who knows, itā€™s all a gamble!

3 Likes

When I talk to people about biohacking I have taken to doing a polypharmacy check with chatGPT. I find it quite useful. For example:

what are the interactions between pitavastatin, empagliflozin, telmisartan

ChatGPT said:

Great question! Letā€™s break this down clearly. You are asking about pitavastatin, empagliflozin, and telmisartan, and how they interact if used together. Hereā€™s a detailed but simple explanation:


1. Pitavastatin (Livalo)

  • Class: Statin (cholesterol-lowering)
  • Main use: Lowers LDL cholesterol and triglycerides, raises HDL.

2. Empagliflozin (Jardiance)

  • Class: SGLT2 inhibitor (diabetes & heart failure med)
  • Main use: Lowers blood glucose, helps in heart failure and chronic kidney disease.

3. Telmisartan (Micardis)

  • Class: ARB (Angiotensin II Receptor Blocker)
  • Main use: Lowers blood pressure, protects kidneys (esp. in diabetes).

:white_check_mark: Possible Interactions and Considerations

Drug Pair Interaction/Concern Clinical Significance
Pitavastatin + Telmisartan Minimal direct interaction. Slight theoretical risk of increased statin level via OATP1B1/3 inhibition by telmisartan, but not usually clinically significant. Generally safe. Monitor for statin-related muscle symptoms.
Pitavastatin + Empagliflozin No significant interaction. No known effect on each otherā€™s metabolism. Safe combination.
Empagliflozin + Telmisartan Additive effect on lowering blood pressure. Risk of hypotension and dehydration. Both affect kidney function (monitor eGFR). Monitor BP, volume status, and renal function.

:warning: Key Monitoring Points

  1. Kidney function (eGFR, creatinine):
  • Both empagliflozin and telmisartan affect kidney filtration. Monitor regularly.
  1. Blood pressure:
  • Both telmisartan and empagliflozin lower BP. Risk of hypotension, especially in elderly or volume-depleted patients.
  1. Muscle-related symptoms:
  • Statins can cause muscle pain/weakness. Though interactions are low risk, monitor if telmisartan affects pitavastatin clearance slightly.
  1. Electrolytes:
  • Telmisartan can increase potassium. Monitor for hyperkalemia.

:white_check_mark: Conclusion: Safe Combo with Monitoring

  • Pitavastatin, empagliflozin, and telmisartan can be safely used together with appropriate monitoring for:
    • Blood pressure (risk of hypotension)
    • Kidney function (risk of reduced eGFR)
    • Potassium levels (risk of hyperkalemia)
    • Muscle symptoms (rarely due to statin)
2 Likes

Horses for courses. I find chatGPT here entirely useless. Look at that response/analysis. Have you ever seen anything quite as superficial and pedestrian? It is the laziest most simplistic ā€œanalysisā€ possible. Itā€™s entirely in character with what I found before. Some weeks ago I asked chatGPT for possible health impacts of simultaneous use of rapamycin and pioglitazone. I was appalled by the level of ā€œanalysisā€ vomited back. As in: no analysis, instead, it was the most childishly simplistic ā€œstackingā€ of possible effects, without actual biochemical analysis that might underpin such interaction. It just takes the effects of the medications, makes a list, and where it sees a duplication, it assumes a stacking of the effects, which we know is rarely the case. And thatā€™s the extent of the ā€œanalysisā€. Utterly, utterly useless. Zero attempt to probe the biochemistry for complex possible interactions or thorough search of the literature with a nuanced evaluation of each paper. There is no knowledge based inquiry here. Itā€™s the reading of the drug inserts for intended indication and side effects and then drawng a straight line between them. To say that I have the deepest contempt for the capabilities of chatGPT in this realm is to pay it a compliment. I find the hysterical hype surrounding this technology to be entirely off putting. YMMV.

2 Likes

a) I find it better than nothing

b) I found ChatGPTs response on melatonin in the third ventricle of the CSF very good. It was quite hard to find this by traditional search even when I had the response from ChatGPT

but

YMMV

I think there is an advantage to doing a polypharmacy check using ChatGPT compared to not doing one at all or relying on my own knowledge. Doing it by a cursory google search may or may not help.

1 Like

Try keep big pharma profit in mind and then you will understand.

1 Like

I asked this in an AMA with Kaeberlein, he seemed more bullish on rapamycin, but said we donā€™t really know:

30:17-35:51

5 Likes

Thanks for sharing. So yeah, pretty much a similar message: Rapamycin maybe working on a more fundamental level, and SGLT2i working on metabolic functions and thus impacting diseases which depend on that.

I think I understand ok. I donā€™t think thereā€™s any sort of conspiracy to try and hide beneficial effects of Rapamycin. I just think that itā€™s difficult to find somebody to pay a huge amount of money for a trial when thereā€™s no profit to be made. (Hence needing some benevolent, curious, rich person)

I think this makes the most sense. As you say, things like CVD, pre-diabetes have massive impacts on longevity, especially if you look at it from the probability point of view. Thankfully, we can pretty much neutralize them with modern medicines. That comes back to the original point that maybe Rapamycin is working on something more fundamental about aging itself. As you said, it seems pretty safe overall. I started a few weeks ago, and Iā€™m working my dose up to 6mg/week. Hopefully Iā€™m still here when marmoset, dog etc trials come up, and I can adjust accordingly. As you say, itā€™s a risk, but at least a calculated one. Gotta keep this in the context that some people literally smoke 40 cigarettes per day and havenā€™t done any sort of risk:benefit analysis on that haha.

1 Like

@mondagai & @CronosTempi: The big problem is that the data around rapamycin is mostly based on organ transplant and cancer treatment which are different dose protocols which the longevity community uses. But if we look at the other compounds then those doses are more well studied and not super high doses. Now when the patent is no longer valid for rapamycin then the incentive to conduct expensive clinical trials is very low because there is no money to earn on this regardless of how good rapamycin is. This is very sad but that is how things are and therefore most likely Rapamycin will never be FDA approved for longevity and an off label drug. One thing that may be FDA approved in the future is some type of next generation of mTOR inhibitor or some type of combinational therapy with rapamycin. This is something that my lab initiative Rapamycin Longevity Lab is working on to push forward in different ways and on top of this also get a much bigger longevity effect.

@relaxedmeatball Well written! I can also recommend checking out the Joan Mannick I did last year with her. In that interview she goes through quite nicely the historical research she has done and what mTOR inhibitors she is working on now and in what way. She is still convinced that an improved immune system is the right way to go.

2 Likes

Sorry for the off-topic post, feel free to remove it.

So I asked o1 the prompt @John_Hemming provided again and this time included your post in the prompt so it knew how to do better. This was the result:

Below is a more in-depth look at how pitavastatin, empagliflozin, and telmisartan might interact, going beyond the usual boilerplate ā€œdrug interactionā€ summaries. Iā€™ll highlight known transporter and receptor pathways, metabolism routes, and mechanistic considerations rather than just stacking the basic side-effect profiles:


1) Pitavastatin

  • Mechanism & Metabolism
    Pitavastatin is an HMG-CoA reductase inhibitor used primarily to lower LDL cholesterol. Unlike some other statins (e.g., atorvastatin, simvastatin), itā€™s minimally metabolized by CYP3A4. Instead, pitavastatin depends heavily on hepatic uptake via organic anion transporting polypeptides (OATP)ā€”particularly OATP1B1ā€”followed by metabolism through relatively minor CYP2C9 pathways and glucuronidation.
  • Transporter Relevance
    Because pitavastatinā€™s hepatic uptake relies on OATP1B1, any co-administered drug that substantially inhibits OATP1B1 can increase pitavastatin plasma levels and, theoretically, the risk of myopathy. Weā€™ll see below why this is relevant for telmisartan.

2) Empagliflozin

  • Mechanism & Metabolic Pathways
    Empagliflozin is a sodium-glucose cotransporter 2 (SGLT2) inhibitor. Its primary route of elimination is via glucuronidation (e.g., UGT2B7, UGT1A9), with minimal involvement of CYP enzymes.
  • Potential Overlaps & Biochemical Considerations
    • Hemodynamic & Renal Implications: Empagliflozin reduces glucose reabsorption in the proximal tubule and can lead to mild osmotic diuresis. In combination with antihypertensives (like telmisartan), there can be an additive blood-pressure-lowering effect and a slight risk of volume depletion or hypotension.
    • Metabolic Effects: Empagliflozin improves glycemic control and may provide additional cardioprotective effects. Thereā€™s no strong direct effect on the OATP1B1 transporter or pitavastatin metabolism. However, through improved glucose handling and possible blood-pressure synergy, it can indirectly alter overall cardiovascular risk profiles when combined with statins and ARBs.

3) Telmisartan

  • Mechanism & Metabolic Pathways
    Telmisartan is an angiotensin II receptor blocker (ARB) primarily cleared by glucuronidation (UGT1A3, UGT1A7, etc.). It has a notably long half-life among ARBs and unique partial PPAR-Ī³ agonist activity.
  • Transporter Interactions
    • OATP1B1 Inhibition: Telmisartan can inhibit OATP1B1 in vitro and possibly in vivo. This is crucial for pitavastatin, which depends on OATP1B1 for hepatic uptake. If telmisartan meaningfully inhibits that transporter, pitavastatin concentrations could rise, theoretically increasing the risk of statin-associated side effects like myopathy or elevated liver enzymes. The real-world magnitude of this interaction can vary, and not every patient experiences a clinically significant rise in pitavastatin levels.
    • UGT Overlaps: While telmisartan is handled by certain UGT isoforms, it doesnā€™t heavily affect the primary UGT pathways for empagliflozin (like UGT2B7) in a way that suggests major synergy or conflict. Still, as with any shared glucuronidation pathways, thereā€™s theoretical potential for shifts in plasma levels, though usually not clinically profound.
  • PPAR-Ī³ Effects
    Telmisartanā€™s partial PPAR-Ī³ agonism can improve insulin sensitivity. In a patient also on an SGLT2 inhibitor like empagliflozin, this might produce slightly more robust metabolic benefits (e.g., further improvements in insulin resistance). This synergy is sometimes leveraged therapeutically, rather than seen purely as a risk.

4) Putting It All Together: Potential Interactions & Clinical Nuance

  1. OATP1B1 Interaction (Telmisartan & Pitavastatin)
  • The most concrete biochemical interaction here is that telmisartan can inhibit pitavastatinā€™s hepatic uptake via OATP1B1. In practice, this could raise pitavastatin plasma levels. Whether or not that becomes clinically relevant depends on individual patient factors (e.g., genetic variants in OATP1B1, dosing, and overall tolerance). Still, itā€™s the main interaction mechanism worth noting from a ā€œdeeperā€ standpoint.
  1. Hemodynamic & Metabolic Synergy (Telmisartan & Empagliflozin)
  • Both telmisartan and empagliflozin reduce blood pressure, albeit via different mechanisms (RAS blockade vs. osmotic diuresis). Additive hypotension is possible, especially in volume-depleted or sensitive individuals.
  • Telmisartanā€™s partial PPAR-Ī³ agonism can synergize with empagliflozinā€™s improvements in glycemic control, potentially offering expanded metabolic benefits for someone with hypertension, hyperglycemia, or metabolic syndrome.
  1. Overlap in Adverse Effects
  • Renal Considerations: Neither pitavastatin nor empagliflozin nor telmisartan is highly nephrotoxic in standard doses, but in patients with compromised renal function or volume depletion, monitoring kidney parameters (eGFR, electrolytes) is prudent because empagliflozin can reduce intravascular volume and telmisartan can modulate renal perfusion through RAAS blockade.
  • Hyperkalemia: ARBs can raise serum potassium. SGLT2 inhibitors are generally neutral or slightly reduce potassium, so thereā€™s not a strong additive risk here as might be seen with certain other drug combinations.
  • Myopathy Risk: While pitavastatin can cause myopathy, itā€™s considered relatively safe compared to other statins in this regard. However, OATP1B1 inhibition by telmisartan remains a theoretical concern. If muscle-related symptoms arose, that would be the prime suspect mechanism.
  1. Literature & Nuance
  • Published data specific to the triple combination (pitavastatin + empagliflozin + telmisartan) are sparse, but individual pairwise studies (particularly for telmisartanā€“statin interactions and the combination of ARBs with SGLT2 inhibitors) do exist. Clinical vigilance typically involves monitoring LDL response (efficacy for pitavastatin), blood pressure (for telmisartan and empagliflozin synergy), and any emerging muscle-related or electrolyte anomalies.

Summary Thoughts

  • Biochemical ā€œRed Flagā€: Telmisartanā€™s ability to inhibit OATP1B1 is the most direct, mechanistic caution point, potentially affecting pitavastatin plasma levels.
  • Beneficial Synergies: Empagliflozin and telmisartan may offer complementary cardiometabolic benefits, and adding pitavastatin provides further atherosclerotic risk reduction.
  • Clinical Interpretation: Real-world significance varies by patient. In many individuals, these medications can be used together safely with appropriate monitoring. When analyzing possible interactions at a deeper biochemical level, transporter effects (especially OATP1B1) and shared metabolic pathways (particularly for glucuronidation) are what stand out.

Thatā€™s the deeper-level rationale beyond simply listing side effects. It explores the main interplay around hepatic transporters, partial PPAR-Ī³ activity, and additive cardio-metabolic effects rather than purely stacking ā€œdrug insertā€ bullet points.

Thank you @BeatTheOdds, for that. Fairness compells me to admit that the analysis you posted is much superior to the previous one. I suppose the takeaway here is that there is a methodological approach to these AI agents - if you know how to ask questions in ways that the AI agents can leverage their databases appropriately, you can obtain results much closer to your objectives.

In putting together my polypharmacy stack, I of course did a lot of research, trawling through the literature, putting together models and testing them against the PubMed knowledge base. It goes without saying that the results of my analysis are considerably more detailed than the analysis posted above, but it is also a reality that my research took many months. The greatest advantage that I see with these AI agents, at least in present form, is that if you know how to approach them in asking questions, they have the potential to vastly accellerate the process of such analysis. Perhaps I could have accomplished what took me months in mere weeks or even days if I knew how to properly take advantage of AI agents. I suppose thatā€™s something I should take a look at. But my other thought is that the development of these is hopefully fast enough that they might be much more useful in a reasonably near future - and at that point might be ready for prime time.

Going forward, the complexity of interactions scales exponentially with each additional drug in the stack. In particular that I also take quite a few OTC supplements (such as taurine, astaxanthin and other carotenoids, vitamins, minerals etc.) - on more than one occasion I was surprised to see an interaction or at least the potential for one, which necessitated my adjusting timing, dosages or sequences, which really is a reminder that you cannot take anything for granted (f.ex. empagliflozin depletes lithium, telmisartan spares it etc.). This is where Iā€™d hope that a future much more advanced AI with more crunching power could make the process of analysis of such complexity easier - although mathematically this is quite daunting, and the permutations reach uncountable numbers in practical terms with shockingly low number of molecules examined - thatā€™s what makes this so difficult and why I am a little skeptical about just how revolutionary AI will be in drug discovery: itā€™s not a question of better algorithms and more crunching power - this is just pure math and the power of numbers, you go parabolic surprisingly quickly. And when dealing with biological processes the number of factors you need to take into account is enormous - you barely scratch the surface and youā€™re already in uncountable numbers territory, immune to any conceivable brute force calculation approach. And therefore I feel that unless thereā€™s some big breakthrough in biological modelling, AI will hit an uscalable wall surprisingly soon. But, every little bit helps, and weā€™ll see. Thank you again.

Also: mods, if this is OT, feel free to remove this post!

1 Like

because that is precisely what a ā€œlarge language modelā€ does! it cannot think in the way you desire, and it is sad that so many do not understand the big difference

and this is why most humans like these chat bots so much, because they are human in the way they are so good at confabulating, seeming to know, without thinking. they dislike thinking, mostly being repulsed and scared by true thinking

the day will come when AI starts to think, and that will be the end of us as more than legacy modules

After reviewing so many rapamycin (and rapalogues) studies, I admit not finding any human direct longevity evidence.
However, I was quite surprised by the amount of case studies using rapa in various challenging diseases and conditions. Good results! My conclusion for now is that the off label use of rapamycyn should be better known among doctors. (I wrote an article about it that will be published in a Brazilian magazine for doctors with interest in the orthomolecular practice.)
As for me, I placed my rapa in the refrigerator because I am not sure if I want to try just nowā€¦ as I express cystic fibrosis (mucus in lungs/pancreas/digestive tract) and I am dealing with the situation well these daysā€¦ Not a single study of rapa on CF so I think I should keep doing what is working for me and skip novelties for now:)

4 Likes

I felt the pedestrian comment. I dont use chatgtp often but the ones i use Sometimes doesnā€™t answer at all and sounds like its trying to bs through a school question where the student doesnt know the answer or stretch the number of words on the paper bc the teacher had a minimum word requirement for the essay. It says stuff you would have to already know to ask the question in the 1st place. I get better results sometimes by specifying, ā€œwithout fluffā€. Still sometimes anemic for the question i asked, but i dont have to wade through as much.

I dont know if the billionaires have better AI than us, or if as usual (dot.com bubble, web3, etc), overestimate technologyā€™s ability and underestimate the amount of work people do.