This is typical of working with agents, not only maths. "I implemented this with agent A, and don't understand the logic, but agent B think its correct". But then you ask a simple clarifying questions to B on a random thing in the code and it'll tell you that this part is actually wrong. At least in maths they have Lean...
This is what we do in my company, give the guidelines to the model. Unfortunately, the model isn't always following them and it's impossible to enforce them manually. There's too much code to review, the code is generally low quality, and is more often than not auto-approved by AI.
Unless you're working on your own project on which you have full control, or you have a strong authority figure in your company who cares about that, it's a lost battle.
It's sad for anyone who love their craft, but tbh I'm not even sure it matters anymore for most projects.
Either you can put it in a linter and automate/force it, or people/agents will take liberties. I feel like there's usually exceptions where a well-meaning guideline doesn't properly fit a specific scenario.
To give a different point of view: I've been taking Bromazepam for 25 years, but very occasionally when I'm very anxious or can't sleep, a few times a year, never more than 1-2 days in a row. It's been helpful, especially when I was younger as sometimes I was paralyzed by anxiety. I never had the slightest sign of addiction but was also careful not to rely on that too much (and never had any addiction problem).
I've also seen friends or family members who used and stop these drugs without much difficulty.
This isn't medical advice obviously, but I wouldn't rely on anecdotal experience to form an opinion on this. These drugs can certainly be helpful to some people.
Seconding your experience. Used on an occasional basis, benzos can be helpful without resulting in the horror stories related here. Of course, everyone is different and YMMV, but confirming that your positive experience is not unique.
They really are extremely effective drugs when used correctly. That's why the calls to ban them (as seen elsewhere in this thread) are not terribly compelling. There's no replacement for what they do.
That said: do not abuse benzos. They will fuck you up.
I was prescribed temazepam for insomnia. It was the only sleep medication that worked, and it was the best sleep I've had in my life. It stopped being effective after around 30 days (or maybe 60, I forget). That was the end of that. I could have argued for a higher dose or a refill or whatever, but... you don't fuck with benzos.
Benzos can fuck you up good and proper without abuse. I was prescribed some for panic attacks. Take as needed, up to every 4 hours.
I took one a day for 4 days, as the panic would flare up right before I tried to sleep. I figured, I could be taking 6 a day, so 1 a day shouldn't hurt. And it was only 4 days in a row. No problem.
Day 5. Bang. Dt's. Immediate anxiety attack. Suicidal thoughts. Brain zaps. High blood pressure. Vertigo.
They are no joke, but doctors throw them around like they used to with opiates.
>They really are extremely effective drugs when used correctly. That's why the calls to ban them (as seen elsewhere in this thread) are not terribly compelling. There's no replacement for what they do.
>That said: do not abuse benzos. They will fuck you up.
This is pretty much my perspective, except I'd say that it doesn't even necessarily require abuse per se. I'd say just be very careful when using them, and get off of them ASAP. Myself, I only take them if there's a procedure of some sort they're called for.
I'd say awareness is more called for than a total ban. A total ban would be messed up because there is zero replacement for them for what they do -none.
There seems to be a reduction in psychological dependence in people treated with benzos and barbiturates for seizure disorders compared to people who don't have seizure disorders.
I think other disorders in gaba/glutamate functioning (or in an adjacent disorder that affects these systems) could be similar.
My guess is that low doses of benzos used to treat neurological disorders have a lower risk because GABA/Glutamate/related systems in that population are pathologically dysfunctional and benefit more directly from these medications.
This needs the caveat that following the maxim "do not abuse benzos" means telling well meaning doctors no if they suggest taking benzos consistently for even a few days. Which is a bit of a stretch of the definition of abuse.
It seems that you got lucky, which is wonderful but not necessarily actionable for others.
Not quite luck, there was a bit more to it than that. Temazepam has a bit different risk profile than some of the others. And the stop signal was developing tolerance; increasing dose would have been the exact thing to do if I wanted to get addicted. So I didn't ask for that!
No one questions if they're helpful. But the danger and the severity needs to be more commonly known.
I certainly wouldn't want them removed -particularly after spending 4 months going through different surgical procedures where benzos were medically indicated. But awareness of the down-side is thin on the ground (from my perspective) and it's good to get out there.
My involvement in inpatient drug abuse clinics has been benzo addiction is one of the hardest to treat and accounted for nearly half the patients, the rest split between alcohol and opioids, even during the height of the oxy crisis. It’s much less reported on because it’s widely prescribed and it’s nearly impossible to OD on. The effects aren’t as dramatic as alcohol, but are no less debilitating. Furthermore because benzos and alcohol have similar pathways they are often abused together and they amplify each others effect. You can satisfy the addiction with either - so people withdrawing from alcohol are often given benzos and people addicted to benzos often drink to relieve their addiction. People describe klonopin and similar long lasting benzos as a six pack that lasts 18 hours - which is a long time to have than level of declined inhibition. This further leads to more drinking, but also to other out of character behavior like overspending - some patients have bankrupted their families through excessive and bizarre spending.
Many people receive opioid treatment and do not become addicted. In fact most do. Most people can drink alcohol without becoming an alcoholic. This isn’t a proof that “anecdotal experience” that opioid addiction and alcoholism aren’t serious problems. In fact it is a demonstration that they are insidious for that exact reason - it’s fairly easy to slip through the cracks in the system into a serious problem that is life altering and extraordinarily difficult to emerge from.
Finally, the hardest withdrawals I’ve seen are benzo withdrawals. Relapse is also extremely high because the profound anxiety that led to the addiction is amplified in the years of subsequent withdrawal with very little as effective as a benzo to mitigate it. Paralyzing irrational fear is in many ways worse than pain, and pain is often a symptom to boot.
I feel that perhaps saying “well not everyone, take my anecdotal experience” isn’t the most empathetic response, and the statistics bear this out. The prevalence of benzodiazepine dependency is roughly on par with alcoholism in the US and several times that of opioids. The distinction is that crisis level addiction is roughly the same as opioid, while alcohol is roughly 7x worse. This likely because acute alcoholic behavior is rapid and very readily observable, while opioid and benzo dependency can be masked behind an otherwise normalish appearing life. The decline is fairly slow often unless you trip over into fentanyl. Further the prevalence of common use of both opioids and alcohol is much higher because surgery and recreational alcohol use is fairly prevalent while anxiety disorders less so.
Stevie Nicks is a popular icon who has been very public about how hard her withdrawal from Klonopin was, including how her skin sloughed off among other horrible effects. She lost a large part of her life span to it. But otherwise it’s also a relatively under reported dependency given its prevalence.
I’d also note in many countries benzodiazepines are highly controlled with much higher bars than US, including the UK, where they’re only allowed in crisis, not general anxiety disorders less.
I also take it on as needed basis. .5 mg clonzepam before bed. Sometime I’d take 2 if the day has been especially rough and I need a good night sleep. It mostly help with falling asleep and feeling more centered the day after.
During a crisis I took it for a month. Then stopped. Stupid as it may be, I only fairly recently learned that this drug can cause hell but for the life of me I couldn’t understand how. All it did for me is help me sleep. I didn’t understand how it can be abused or what’s the recreational part of it. When my doc told me about the dependence, and how it may creep stopped even my occasional use and the bottle I got last Jan still sits in my meds canbinet.
Could it be that different people have different reactions or may be more susceptible to addiction to it?
Maybe there are different types of downside; and different people have different experiences?
For example, if I take some sort of sleep aid (e.g. US OTC antihistamines, Ambien, or low-dose benzos) then after ~2 nights my subsequent ability to fall asleep without help is noticeably impaired. That’s already the start of a physiological change that, if not reversed, would continue to a dependency (albeit not necessarily an ‘addiction’.)
It’s a process of analogue change, not binary, and individuals’ pace of change will vary.
I have GAD and ten years ago when I was on my lowest, I was on 2 mg klonopin daily plus SSRI.When I asked to wean off the klonopin, we did over the course of two months and zero problems (fortunate).
Very similar. I had daily panic attacks for a while. Lorazepam and therapy helped me recognize the signs and slowly they went away.. along with how often I even took the lorazepam. It’s useful every now and then but it’s very rare.
Some people are wired for addiction and some aren't. I knew a guy who did heroin or cocaine a few times a year when he felt like it and never got addicted. I also quickly lose interest in alcohol or marijuana but I can't stop eating sweets no matter how hard I try...
People working at Meta tend to gravitate around the same group of tech companies who offer the same range of salaries, same selection process, same internal structure. Many of them aren't particularly brillant SWEs (in the sense you don't need to be an elite coder or top PhD), although they can deliver based on the evaluation processes in these companies.
Meta has a bad reputation in some circles, but I don't think there's anything peculiar or particularly evil about them, at least not worse than similar publicly traded big tech companies. They are amoral money-making entities.
That being said I agree that their products suck. I don't use them, besides what's app because it's difficult to avoid. It's also end to end encrypted so it doesn't really matter regarding mass surveillance.
> It sounds and feels a lot more like an engineering or product management role—something a lot of programmers would never in a million years have signed up to do voluntarily.
True, but dealing with agents is still easier than dealing with people (at least for introverted personalities). That being said, you still need to deal with other people's army of agents.
> Especially the toxic manosphere culture at the moment.
The experience of living/traveling in the US is very different from what can be seen on social medias and forums. At least in big cities in blue states.
> The idea/justification is that Claude is given the ability to voluntarily stop its conversations, thus any token it outputs that isn't a stop token is voluntary.
I don't get this argument. How is it different from any algorithm that returns a stop token on some inputs? what makes the token "voluntary" for Claude?
> The fact that aliens have not nuked us from orbit out of sheer panic is
Why would they care? maybe they're aware of millions of inhabited planets and life isn't anything special for them.
Also, humans are a dominant and invasive species right now, but in geological times, our dominance couldn't be nothing more than a spark. Scenarios for destruction of civilization are numerous (nuclear war, pandemics, climate change, super volcanos, asteroids) and very possible even in the short term. Trump could do it tomorrow morning if he feels like it.
reply