Thats still pretty bad. Its no longer private if all your code goes through LLM training set and is resurfable to everyone publicly.
Why would I ever use copilot on any code Id want to be kept private? Labling it a private repo and having a tiny clause in the TOS saying we can take your code and show it to everybody is just an upright lie
I mean, you shouldn't send data to any SaaS LLM for code you want to be private, unless you have had them sign some sort of contract saying they will not train on your use. In fact, it is probably never a good idea to send anything you want to be private off premises unencrypted.
I dunno how you guys even go throuh the $200 subscription. I use it every day for work and side projects doing tasks in parallel and Im no where newr the limit on $100.
There's definitely use cases for this for long running tasks, like doing research, but for typical use cases they require way too much constant supervision and interaction
If you take it and you feel your anxiety is lessened, that's the greatest proof you can ask for. All the psychiatric studies are already based on self assessment.
Second, a lot of psychatric treatments are temporary, ending whenever the medication is stopped or wears off so I dont see how this would be any different
> If you take it and you feel your anxiety is lessened, that's the greatest proof you can ask for.
This mistake has been made many time throughout history. Cocaine was originally believed to be a viable treatment for depression. Opioids and amphetamines too. You take them and you feel good for a while, which was mistakenly equated with treating depression.
Many drugs will make you feel good temporarily by blocking certain feelings or tricking your brain into feeling good. This is not the same as treating a condition.
You can think of actual treatments as working closer to the source to reduce the problem, not temporarily overriding it with a powerful drug-induced sensation.
Psychiatry as its practiced has no idea as to what depression even is under the hood. The entire science is based on the patients self reported feelings or the psychatrists feeling of how someone else is feeling.
What you're saying is something else, that drugs can produce long term harm despite short term improvements
>> This study is showing that THC, CBD, blends, or cannabis do not improve self-reported feelings over the long term.
Just to clarify, the study is not saying that.
The study is saying "there isn't conclusive evidence at this point, but it leans more toward helping than hurting on many categories". Please read the paper if in doubt.
I did. I don’t know if you did, though, because the categories that they said there might be some signal were secondary topics like insomnia, not depression PTSD and anxiety.
I mean, it’s literally in the title. It’s covered again in the abstract.
I dont have a problem with the study or its conclusions, just the parent post I was replying to.
>Recreational drugs make you feel good temporarily. That's literally why people do them.
The point Im making is this is true for a ton of psychiatric or even non psychatric treatments. And to be perfectly clear Im not saying you should be treating your anxiety with weed, even if it does help you
> The point Im making is this is true for a ton of psychiatric or even non psychatric treatments
That point wasn't intended to be taken in isolation.
I was making statements about how long-term treatment of an underlying condition is not the same as taking a drug which temporarily masks a problem or induces altered mood states.
The fact that a drug has acute effects, good or bad, is separate from any conversation about chronic effects.
For depression, anxiety, and other conditions it's the chronic effects that matter. The acute effects almost become side effects at that point. For some drugs, getting to long-term treatment involves tolerating the acute effects while your body adapts
You keep ignoring the fact that what you just said applies to current medications used to treat depression and anxiety. They do not treat the underlying issue long-term, and if you stop them, you are worse off than before due to rebound effects, and even if those effects subdue, your depression and anxiety returns. And just to add to this for clarification, antidepressants may treat depression, but it does not cure it either.
Same with amphetamines for ADHD. And yes, if you take much more, you will experience side-effects ("cause harm when abused"). Opioids are not an outlier at all.
> Recreational drugs make you feel good temporarily
Drugs are only recreational if you take them recreationally, there is nothing that makes them inherently recreational.
And we have not discussed MDMA, which is considered a hard "recreational" drug, yet there are lots of benefits for treatment of PTSD, for one, similarly to psychedelics.
... or ketamine for depression, which is now approved by the FDA, even.
The difference between your positions is not about acute vs chronic, it's about tolerance. If a drug for a long term condition has short term effects the first few times and then they fade under regular use, it's less of a valid treatment. Especially if there is a withdrawal effect, and any negative side effects of regular use.
We absolutely overprescribe a lot of psychiatric meds that do not have significant beneficial long term effects. "Stabilizing" a patient in an inpatient hospital psych ward may as well involve a Magic 8-Ball picking the particular antipsychotic for its short term effects, while on the other hand doctors and nurses put people on Seroquel at the drop of a hat in reported sleep problems, and don't take them off until natural death or until the essential tremors get reported decades later.
This is a fair point but there’s also truth to the fact that “I feel better” does not automatically mean it’s an effective/good treatment.
Lots of depressed people turn to alcohol to provide a mood lift. But you can’t be drunk all the time and function, and when the alcohol wears off you feel even worse. So it’s a terrible treatment.
People talk a lot about cannabis like it’s a great treatment for all kinds of stuff. But is it closer to a drug you can take on a schedule that boosts your mood essentially all of the time with acceptable side effects? Or is it closer to a drug that lifts mood very short term and then makes it worse?
> “I feel better” does not automatically mean it’s an effective treatment.
Of course, I agree with that.
I also agree that you cannot be drunk to function, but there are many other "drugs" that people would oppose that do not make you dysfunctional.
Personally I would not use cannabis because I know that it makes me dysfunctional, but it may not be the case for some people for all I know.
FWIW I take opioids for my chronic condition, and it also helps with my emotional volatility, depression, and anxiety, too. I have not experienced any side-effects either.
Alcohol is so terrible, especially when severely abused, that "better than alcohol" absolutely cannot be a meaningful standard for being a proper treatment.
> there are many other "drugs" that people would oppose that do not make you dysfunctional.
People oppose these drugs because they do tend to make you dysfunctional, at least when abused. And when people are severely depressed, the depression itself makes it more likely that they will abuse their drugs.
> You keep ignoring the fact that what you just said applies to current medications used to treat depression and anxiety. They do not treat the underlying issue long-term
Those do cause improvement in self-reported feelings long term, i.e. they lower the baseline levels over a long period of time, rather than just for a short period right after you take the drug.
But you'd be right to say that they're not very good, i.e. that doesn't help your life very much. If there's an actual underlying cause, like sleep apnea, treating that will help a lot more.
Ketamine is a harder case, it really does cause improvement, but it lasts about two weeks. It also can cause psychosis, which is very dangerous. The s-ketamine the FDA approved for use in treatments is also via an inhaler, so it's both purer and via a different route than other a lot of other ketamine usage and it was approved because it actually showed a measurable effect in studies.
But it's really awful to use and if you find out that, say, sleep apnea was actually causing the issues, treating that will do a lot more good than inhaling s-ketamine ever did.
As you may have inferred, I write this based on personal experience.
The problem with your argument is it's whataboutism. Your argument's conclusion should be that even prescription drugs aren't necessarily good.
Really what is wrong is that most prescription drugs do show less tolerance. Yes, prescription drugs have tolerance, but not as fast as recreational drugs taken at recreational doses.
By giving a patient the ability & skills to establish a less dopamine-seeking lifestyle while temporarily relieving them of the deficit.
Its use is supposed to be coupled with therapy and/or coaching (e.g. https://www.thriveemerge.com) to ensure that the patient isn't just using it as a lifelong crutch.
That's how it's supposed to be done. This approach is more effective in children for obvious reasons. Persons diagnosed later in life are therefore more likely to require it permanently.
Why do you need amphetamines to do this? Why not a legal stimulant? Even something as trivial as coffee can be effective if you take it strategically, i.e. stay off of it completely until the rare times when you need the turbo boost.
There's also atomoxetine, but it's not very effective.
> Why not a legal stimulant?
Which ones? People absolutely do self-medicate with coffee, ephedrine, or even cocaine where it's available (coca tea). And these stimulants do work, but they have _more_ side effects than amphetamines when used in theurapeutic doses.
Yeah, that's what I did. Back before I started ADHD treatment I was averaging about 12-14 cups of coffee a day, and that's before chocolate. That was the only thing that ever calmed me down. My heart pain eventually got so bad that I had to switch.
Would love to learn more about how this is the case (both assertions). Mind sharing?
Neuroscience seems to be coming through with more and more understanding using technologies like fMRI and others the past 5-10 years. There is definitely some understanding there.
The short version, as I understand is, is that brain scans show differences at the population level but not on the individual level. Amen claim to both diagnose ADHD through brain scans (which is already impossible) and also diagnose various "subtypes" of ADHD like "Limbic ADD" which have no scientific backing for their existence.
You can pirate a copy of the DSM-V and see the diagnostic criteria for yourself. It's particularly interesting to me (as I believe the field has "crystalized" more than it has right to) that two people can be diagnosed with "ADHD" but share few overlapping symptoms. Separately, fMRI evidence is still not solid on its own, it has to be paired with stronger science to be at all useful, but even so studies based on it tend to suffer from numerous problems. The most infamous case being a "study" that found brain activity using fMRI of a dead fish. That was in 2009. In 2020, Botvinik-Nezer et al. published a paper about 70 different research teams analyzing the same fMRI dataset and producing wildly different results. It's a blunt instrument but people are deceived into thinking it's really solid; it's not.
Insightful, thanks. I typically separate the personality from the assertions and look at the assertions individually in present time due to the incredible lag time between knowledge and it becoming practiced in the clinic. Professionals can best practice 5-10-15 year old information because they haven't caught up, there's a lot to stay on top of.
The first link is from 2008 and claims EMDR doesn't work (I believe it emerged in the 90's and new things are generally resisted in health until they're not), and that Hyperbaric Oxygen Therapy doesn't help the brain, both are understood to be helpful in ways beyond this article in 2026. Unrelated to Amen, fMRI is helpful with TBI imaging.
not clinical understanding, not in any useful way. its another tool for study, but my understanding is that aside from some very high level structural information, there's sadly not that much to be learned from watching the flashing lights.
fMRI just measures the BOLD signal (blood oxygenation). It's a proxy for neural activity, meaning any conclusions drawn from it are inferences, not direct observations of the brain 'thinking'.
> Cocaine was originally believed to be a viable treatment for depression.
Is it not??
Sure there's the addiction and harm from abuse that make it less than ideal for long term use, to put it mildly, but weed isn't coke so what's really the argument here?
As someone who has had depression literally as long as I can remember, being able to releve my symptoms when I really need it, even for just an hour, would be life changing.
Duration of effect matters when it comes to successful treatments.
If we take your position and apply reductio ad absurdum, we could say that cocaine is a highly effective treatment for anxiety, although of course we know that in the not-so-long run it has the opposite effect.
But a lot of psychatric treatments are just that. Treatment for ADHD for example is giving ampethamines (which btw are chemically no different than a low dose of meth), which have a duration of 3-6 hours and its back to worse than baseline after the effect has worn off.
There are multiple treatments for ADHD, including alpha-2 receptor agonists and norepinephrine reuptake inhibitors. Some of them show patterns of increasing efficacy out to a year (the length of the study).
The reason amphetamines are used for ADHD but not depression is that they've been studied to show that the ADHD improving effect can remain for many months, while the mood-improving effect will taper off quickly if you take them every day. Almost everyone who takes ADHD stimulant, feels a mood and motivation boost ("so happy I could cry" is the common phrase) and then is disappointed when that mood boost stops happening after a few weeks or months will learn this. Attention enhancement is less prone to tolerance, though it still accumulates tolerance too. There are some studies showing that the effects of stimulants in ADHD diminish substantially on a multi-year time frame, and it's probably not a coincidence that many people (though not all) who take stimulants discontinue after several years.
No not just like it, because the only difference with methampathemines is that the added meth group makes it able to cross the blood barrier much quicker, hence why I said its equivalent to a lose dose of meth. The chemical/biological response on the body and brain are very similar, the difference is in potency
But onset of action is a very important distinction in medicine/pharmacology, as is dose.
Most abusers of methamphetamine are not taking it orally (slow route of administration) and are generally using much higher relative dosing than ADHD patients are using amphetamines. Potential for addiction and other physical harms are greatly affected by both of those things, so the comparison has some truth, but is obviously sensationalized.
No, the difference between amphetamine and methamphetamine is that they are literal different chemicals.
If one could 'add meth'(??) to chemicals to make them more potent, without changing the chemical, it would be the difference between (for example) citric acid and really strong citric acid, or codeine 2.5mg and codeine 5mg.
You'll note that neither of these involves changing the name of the chemical, because that is not how chemical names work.
As someone else has pointed out, the difference between 'hydrogen monoxide' and 'dihydrogen monoxide' isn't 'it's like hydrogen monoxide with added di', because that is ridiculous.
Please stop saying anything beginning with 'meth' is just meth with added bits.
It's a really odd misinterpretation of the terrible dangers of: methane, Methodists, methanol, Methaemoglobin, methicillin, etc.
Well, it's not that simple. It's reasonable to expect that you could see some increased level of oxidative and excitotoxicity. It's harder to draw a bright line around the dopaminergic system specifically because some level of neuronal death is expected over the course of a lifetime. We lose 5-10% starting with middle age yet don't tend to show parkinsonian symptoms until 60-80% are gone.
It's pretty reasonable to expect reversing DAT and inhibiting VMAT2 increases oxidative flux, the question is really how much not if. Methheads certainly get "brain damage", but is nudging the average loss from 5-10% to 7-12% "damage"? Is it meaningful? Over 30, 40 years that could very well add up.
Meth is also used as an ADHD treatment. I think the reason is just the dosages that are used by addicts compared to people who just need the ADHD treatment.
A typical legitimate therapeutic methamphetamine dose is around ~20mg (up to maybe 60mg a day). A typical dose used by addicts is around 1 gram. And it's usually smoked, resulting in immediate bioavailability.
That's a terrible oversimplification. Stimulant treatments for ADHD are not supposed to produce pronounced mood-enhancing effects. Stimulant treatment has been shown to be effective indefinitely in majority of people without increasing the dosage over time.
These days formulations like lisdexamfetamine and extended release methylphenidate are preferred because they have all-day efficacy with typical duration of action of around 8-12h which carries lower abuse potential.
extended release are just two doses of the drug where half the beads are delayed by ~4 hours. How is that different from taking two edibles a day and claiming full day efficacy?
That's not the case for lisdex nor Concerta methylphenidate. Some generics work that way and they're generally regarded as being worse than Concerta.
The benefit is that the medication automatically produces a smooth effects profile allowing you to live your life without timing medication to perfection.
A pronounced come-up and crash is a risk factor for abuse and addiction, so smoothing or removing the peaks and valleys is important.
I mean the difference is that you just take one in the morning? Which makes adherence easier, makes sure that the delay is constant rather than variable, and reduces abuse liability.
(As an aside, there are more complex extended release mechanisms than just delayed bead release - like lisfexamfetamine is a inactive prodrug, so cleaving the lysine off the amphetamine is rate limited. This has the effect of extended the duration of effect, and reduces the potential to abuse by snorting/iv/etc).
Shouldn't that be up to bodily autonomy? If someone feels that cocane relieves their symptoms then who is the doctor to say that they don't. Perhaps releaving those symptoms even for a short period of time is worth the consequences.
That's just arguing for drug legalization with extra pseudoscience.
I am all for people doing however much cocaine they feel they need. In broad daylight - let's just drag that into the light and let people go to the dispensary for cocaine hydrochloride, metered, measured and with warning labels.
Because the war on drugs is a stupid waste of time and lives, but no doctor or medical professional has to justify your own stupid actions.
It’s also an argument for quackery and snake oil, as long as the salesman declares “some people said it works!”
“People should be allowed bodily autonomy to take whatever chemicals they want” easily and dangerously turns into “People should be able to advertise and sell miracle cures that don’t work as long as their victims are gullible.”
Every snake oil fraudster hides their fraud behind principles like bodily autonomy.
There is already much snake oil sales and marketing going on, it's already enabled by arguments that are not based on bodily autonomy, which suggests that a move towards more respect to bodily autonomy will not noticeably increase such snake oil. The resistance to homeopathy has not gone well. I actually believe the resistance is actively harmed by making legit chemicals harder to get instead of easier. When people can easily get the good stuff that works there's not much of a market for the easy to get snake oil that doesn't.
I think the problem is that, at least in my experience, you end up with more anxiety once the initial high wears off. Paranoia is an extremely common side effect of Marijuana, and so are nightmares with prolonged use. And once you kinda get into a routine with it, you have a hard time quitting, because your overall anxiety is raised, and you need it just to get back to a normal functioning level. My guess is that this is due to the effects that THC has on blocking your REM sleep. Without the proper REM sleep, it seems pretty common to be anxious and foggy-brained.
THC cannot both ”cause nightmares” and ”block REM sleep”, because nightmares happen during REM sleep.
People who suffer from nightmares may benefit from less REM sleep. It’s much more refreshing to sleep in a non-REM state and wake up normally than it is to repeatedly wake up in cold sweat and be afraid to go to sleep again.
Doing ten shots of tequila is a 100% scientifically proven cure for social anxiety then. If you take it and your anxiety lessens, that's the greatest proof you can ask for! Let's just completely ignore the crippling morning hangover and liver damage
Sure but my point was that you cannot have an argument where you go "yeah this happens so it's good" while you disregard everything else (it's a different question on how bad it is compared to alcohol etc). But if we follow the logic of the original comment, then it's valid logic since "hey it works so its scientifically proven!" You can replace alcohol with something else to highlight that as well, like how putting out a kitchen fire with a bucket of gasoline is a good idea. It completely covers the flames for a split second! Why worry about the explosion that happens immediately after?
So my comment wasn't about alcohol vrs cannabis but rather how that kind of logic is short-sighted and faulty.
I mean... Yeah. Alcohol is very well documented and even more widely used for exactly this purpose BECAUSE it works.
The side-effects are often terrible. This is also true for many widely-prescribed drugs, and has been even more true in the past. The folks I've known on MAOIs were pretty wrecked.
There’s a difference between intoxication and treating the chemical imbalance behind depression or anxiety. For one thing, treatments for anxiety only target the anxiety: they don’t impair the person the way that weed or alcohol does. (They can have other side effects, of course.)
Drugs for anxiety treatment do wear off, but not the same way that weed or alcohol does: something like Celexa takes a few weeks to build up in the system, and don’t lose effect 12-24 hours later if you miss a dose. I’m not sure how long you’d have to stop before it loses efficacy entirely.
I’m not Nancy Reagan, though: I would not advise people to self-medicate with booze or pot if they’re suffering from depression or anxiety, but I’m not going to preach at anybody who is doing so and thinks it’s working for them. I will say that I’ve seen that end badly, though. I can think of three people I’m close to who’ve tried it and have had problems with addiction: all of them are now sober and (I believe) on regular antidepressants.
Im prescribed adderall for ADHD. It is a high. You feel more positive, more productive, more forward looking for a few hours and its back to baseline or worse when you crash.
As for impairment, it really depends. If weed removes your anxiety and lets you relax, its benefit could be greater for what youre doing than the impairment it causes. And adderall, SSRIs can cause impairment of sorts too.
I won't call you an idiot, but assuming that all cases are simply people being sad is a bit simplistic.
Some people are going to be more susceptible to depression, for whatever reasons, and improving someone's surroundings is probably going to prevent or alleviate depression to an extent, but to people who are depressed now it's somewhat pointless advice.
Psychiatric treatments return a person to a baseline that can be managed with therapy or healthy coping mechanisms.
Chasing a high is not a treatment, it merely defers the problem. As tolerance to the high builds, patients lose the therapeutic value but have gained crippling dependency and addiction.
All that stuff would take me about 5 minutes without AI. Those are things with 10,000 examples all over the web. AI is good at distilling the known solutions. But anything even slightly out of the ordinary, it fails miserably. I'd much rather write that code myself instead of spend an hour convincing an AI to do it for me.
Simple npm install, all of it has already been distilled into dozens of similar repos. Just pick one, install it, and follow the simple use case. 5 minutes if we're in a race.
>done Setup websockets
If this takes you more than 5 minutes, then you're a shit developer.
>stream audio from mic
Again, another npm install or two, simple POC could take 5 minutes.
>transcribe with elvenlabs
I don't know what elvenlabs is, nor do I care, but I doubt it's as complex as the OP thinks it is considering the rest of their comment was about simple, solved problems.
Haven't tried Claude for this, but I can't think how it could possibly do.
I built a game bot using Win32 API to send input and screen capture to OCR and some OpenCV to recognize game elements. Dead simple and actually quite boring and repeatitive after I worked on it for a while. How could Claude agents possibly do this ?
I did use Claude to refer docs and API, though.
That actually sounds like something Claude could do pretty easily.
Yegge's book describes his coauthor's first vibe coding project. It went through screenshots he'd saved of youtube videos, read the time with OCR, looked up transcripts, and generated video snippets with subtitles added. (I think this was before youtube added subtitles itself.) He had it done in 45 minutes.
And using agents to control other applications is pretty common.
My elementary schooler did this with pictures of his stuffed animals last week. I helped a little bit, but most of it was Claude. He's never coded before.
Yes, I've doomed him all because of a 30 minute interaction. Just like when he watched Kerbal Space Program videos on YouTube he lost all motivation to get to the moon himself. Oh wait.
And he definitely doesn't make up missions using the mission builder using if / then loops. He'll never learn to code. Oh the humanity.
I'd rather have my kid typing on a real keyboard into Claude, asking questions about what Python, and modifying the Claude-generated code than watching random videos and playing Roblox on his iPad.
Yes, because search engines are populated with SEO-optimized LLM-filled articles that say nothing of value anymore. The only reason AI-assisted tools are "better" is because Web search is so much worse.
It's like everyone forgot that the first result for anything web-related would be W3schools, and the next 5 would be spam message boards that tries to scrape all the other boards and sends you to a porn site when you click on it.
Yeah it is worse now but I don’t remember it ever being good. If you know where to look and have a trusted set of resources curated sure, but of course you won’t for unfamiliar territory which is exactly what LLMs help with.
I haven't read wikipedia in a long time so I can't answer your question, I am just pointing out that just saying "the facts are correct" is not enough to say there is no bias on wikipedia
Why would I ever use copilot on any code Id want to be kept private? Labling it a private repo and having a tiny clause in the TOS saying we can take your code and show it to everybody is just an upright lie
reply