It likely just hallucinated the ADHD thing in this one chat and then made this up when you pushed it for an explanation. It has no way to connect memories to the exact chats they came from AFAIK.
*not entirely sure. I t seems to frequently hallucinate the address
and your reasoning for this is what?
The point is ChatGPT gets various info about you and it won't disclose to you that it has them.
There's the memory feature, but various reports (and my own experience) indicate that even if you disable it, some stuff you've said before (or the LLM inferred) is still fed into its sytem prompt.
We also know that AI can sometimes make up stuff. I think it might have "guessed" the user has ADHD, this got added into the system prompt and it won't be revealed to the user considering how this works. It wasn't done on purpose and wasn't malicious.
If you want chats to shared info, then use a project.
Yes, projects have their uses. But as an example - I do python across many projects and non-projects alike. I don't want to want to need to tell ChatGPT exactly how I like my python each and everytime, or with each project. If it was just one or two items like that, fine, I can update its custom instruction personalization. But there are tons of nuances.
The system knowing who I am, what I do for work, what I like, what I don't like, what I'm working on, what I'm interested in... makes it vastly more useful. When I randomly ask ChatGPT "Hey, could I automate this sprinkler" it knows I use home assistant, I've done XYZ projects, I prefer python, I like DIY projects to a certain extent but am willing to buy in which case be prosumer. Etc. Etc. It's more like a real human assistant, than a dumb-bot.
I've found a good balance with the global system prompt (with info about me and general preferences) and project level system prompts. In your example, I would have a "Python" project with the appropriate context. I have others for "health", "home automation", etc.
I could not disagree more. A major failure mode of LLMs in my experience is their getting stuck on a specific train of thought. Being forced to re-explain context each time is a very useful sanity check.
Maybe if they worked correctly they would be. I've had answers to questions be influenced needlessly by past chats and I had to tell it to answer the question at hand and not use knowledge of a previous chat that was completely unrelated other than being a programming question.
I remember having conversations asking ChatGPT to add and remove entries from it, and it eventually admitting it couldn’t directly modify it (I think it was really trying, bless its heart) - but I did find a static memory store with specific memories I could edit somewhere.
ChatGPT is just suppose to “work” for the lay person and it just doesn’t quite often. OpenAI is already being sued by people for stochastic parroting that ended in tragedy. In one case they’ve tried to use the rather novel affirmative defense that they’re not not liable because using ChatGPT for self-harm was against the terms of service the victim agreed to when using the service.
If every building I went to in the US had ramps and elevators even though I'm not in a wheelchair, would it be "fucked up" that the building and architects assume I'm a cripple?
There's just as much meaning in ChatGPT saying "As you said, you have ADHD" as a building having an elevator.
In the training data for ChatGPT, the word ADHD existed and was associated with something that people call each other online, cool. How deep.
Anyway, I do assume very single user of this website, including myself, all have autism (possibly undiagnosed), so do with that information what you will. I'm pretty sure most HN posters make the same assumption.
It’s probably a very human trait to do that but it is a bad habit.
So I'm not entirely surprised that an LLM would start assuming that the user have ADD, because that's what part of it's training data suggests it should.
The issue is it doesn't apply here as it's neither a person or a coherent memory/thinking being.
"Thinking" models are basically just a secondary separately prompted hidden output that prefaces yours so your output is hopefully more aligned to what you want, but there's no magic other than more tokens and trying what works.
Say I’m interested in some condition and want to know more about it so I ask a chatbot about it.
It decides “asking for a friend” means I actually have that condition and then silent passes that information on data brokers.
Once it’s in the broker network it’s truth.
We lack the proper infrastructure for to control our own personal data.
Hell, I bet there’s anyone alive that can even name every data broker, let alone contacts them to police what information they’re passing about.
I know in Europe we have the GDPR regulations and in theory you can get bad information corrected but in practice you still need to know that someone is holding it to take action.
Then there's laundering of data between brokers.
One broker might acquire data via dubious and then transfer that to another. In some jurisdictions once that happens the second company can do what they like with it without having to worry about the original source.
I guess unless you have an offline system you are at the mercy of whoever is running the services you use.
Googling allows you to choose the sources you trust, AI forces you to trust it as a source.
The former shows you things people (hopefully) have written.
The latter shows you a made-up string of text "inspired by" things people have written.
ChatGPT may have picked that up and give people ADHD for no good reason.
Did some digging and there was an obscure reference to a company that folded a long time ago associated with someone who has my name.
What makes it creepier is that they have the same middle name, which isn't in my profile or on my credit card.
When I signed up for ChatGPT, not only did I turn off personalization and training on my data, I even filled out the privacy request opt-out[1] that they're required to adhere to by law in several places.
Also, given that my name isn't rare, there are unfortunately some people with unsavory histories documented online with the name. I can't wait to be confused for one of them.
You did all of that but then you gave them your real name?
Visa/MC payment network has no ability to transfer or check card holder name. Merchants act as if it does, but it doesn’t. You can enter Mickey Mouse as your first name and last name… It won’t make any difference.
Only AMEX and Discover have the ability to validate names.
FWIW, I have a paid account with OpenAI, for using ChatGPT, and I gave them no personal information.
They can, and do, validate on street number and/or zip code so you can certainly error out on typos there ... but not name.
Personally, I'm on the fence. I suspect that I've always had a bit of that, but anecdotally, it does seem to have gotten worse in the past decade, but perhaps it's just a symptom of old age (31 hehehe).
The meme of 'ADHD as the "fucked up attention span disorder"' has done immeasurable damage to people, neurotypical and ADHD alike. it is the attribute that is the least important to my life, but most centered towards the neurotypical, or the other people it bothers.
> modern life fucks up your attention span
That said, this statement is true, it's just a fundamental misunderstanding of ADHD as "dog like instinct to go chase a squirrel" or whatever. Google is free, so is Chatgpt if that's too hard.
> I'm not sure it's healthy to want to put a label on everything
I don't particularly care for microlabeling, but it's usually harmless, nothing suggest the alternative of "just stop talking about your problems" is better. People create language usually because they want to label a shared idea. This is boomer talk (see "remember facebook?" no)
> or medicate to fall back on the baseline
I'm not sure "If you have ADHD you should simply suffer because medicine is le bad" is a great stance, but you're allowed I suppose
still one of the most common symptom, and the one everyone use to self diagnose...
> because medicine is le bad
idk man, I've seen the ravage of medicine one people close to me. Years of adhd medicine, anti depressants pills, anti obesity treatments... They're still non functional, obese and depressed, but now they're broke and think there truly is no way out of the pit because they "tried everything" (everything besides not playing video games 16 hours a day, eating junk food 24/7 and never going out of their bedroom, but the doctors don't seem to view this as a root cause)
Whatever you think, I believe some things are over prescribed to the point of being a net negative to society. I never said adhd doesn't exist or shouldn't be treated btw, you seem to be projecting a lot of things. If it works for you good, personally I prefer to change my environment to fit how my brain/body works, not influence my body/brain by swallowing side effects riddled pills until death to fit in the fucked up world we created and call "normality"
Just try harder to make insulin, bro. You can outthink that t1 diabetes if you try hard enough.
This weird macho resistance to and scorn of anyone using the mental tools we have available is why men kill themselves at a much higher rate. There is no award for who struggled the most in life.
Do they always work? No. Do they work for a lot of people? Sure do. Can they be replaced with diet and exercise and different circumstances? Yeah, sometimes. Is that realistic? Not usually.
They aren't magic. They don't forcibly make you happy or change your personality or make your problems go away. You still have to do the work. But they're a crane to help lift the crushing weight you're under so you can shimmy out from under it.
If you don't want to use them, fine, but not using them doesn't make you a better person.
I, and pretty much everyone I talked to who was diagnosed with adhd in late teenage years on, actually primarily used executive dysfunction to self diagnose and then persued official treatment. you pulled that straight outta your asshole, shove it back up there
> doctors don't seem to view this as a root cause
how the hell would doctors know? bring this to any psych and they'd immediately tell you to knock that shit off, but what the hell else are they supposed to do?
> I never said adhd doesn't exist or shouldn't be treated btw
your opinions seems to boil down to "lmao just try harder", which seems less useful than throwing pills at the problem. you have the mental model of a cat
> proooojecting
the words of someone who can't articulate their position
what would you recommend if one is against the idea of medication in general for neurological issues that aren't deterental to ones life?
do you feel the difference between being medicated and (strong?) coffee?
have you felt the effects weaken over time?
if you did drink coffee, have you noticed a difference between the medication effects weakening on the same scale as caffeine?
is making life easier with medication worth the cost over just dealing with it by naturally by adapting to it over time (if even possible in your case)?
this is a personal pet-project of observing how different people deal with ADHD.
Given that ADHD people tend to commit suicide 2x-4x times more often than general population [0] keep in mind that it's not detrimental until it suddenly is.
Also it gets worse with age, so it's better to get under doctor's control sooner than later.
> is making life easier with medication worth the cost over just dealing with it by naturally by adapting to it over time (if even possible in your case)?
I am now 20, admittedly "early" in my career. Through high school and the first 2 years of university I have banged my head against ADHD and tried to just "power through it" or adapt. Medication isn't a magic bullet, but it is clear to me at least now that I am at least able to rely on it as a crutch in order to improve myself and my lifestyle to deal with what is at least for me, truly a disability. Maybe one day I won't need it, but in the mean time I see no reason why attempt #3289 will work for real this time to turn around my life.
ADHD is a debilitating neurological disorder, not a mild inconvenience.
Believe me, I wish that just drinking coffee and "trying harder" was a solution. I started medication because I spent two decades actively trying every other possible solution.
> what would you recommend if one is against the idea of medication in general for neurological issues that aren't deterental to ones life?
If your neurological issues aren't impacting your life negatively, they aren't neurological issues. I don't know what else to say to this. Of course you shouldn't treat non-disorders with medication.
> do you feel the difference between being medicated and (strong?) coffee?
These do not exist in the same universe. It's not remotely comparable.
> have you felt the effects weaken over time?
Only initially, after the first few days. It stabilizes pretty well after that.
> if you did drink coffee, have you noticed a difference between the medication effects weakening on the same scale as caffeine?
Again, not even in the same universe. Also, each medication has different effects in terms of how it wears off at the end of the day. For some it's a pretty sudden crash, for others it tapers, and some are mostly designed to keep you at a long term level above baseline (lower peaks, but higher valleys).
> is making life easier with medication worth the cost over just dealing with it by naturally by adapting to it over time (if even possible in your case)?
If I could have solved the biological issue "naturally" I would have. ADHD comes with really pernicious effects that makes adaptation very challenging.
thanks for sharing, the coffee part is mostly for the claim that it has the opposite effect on people with ADHD or no effect at all.
Unmanaged ADHD is dangerous, and incredibly detrimental to people's lives, but the level of such may not be entirely apparent to somebody until after they receive treatment. I think the attitude of being against medication for neurological issues where that is recommended by medical professionals (including where that for something perceived to not be detrimental enough) is, to say the least, risky.
I would perhaps encourage you to do some reading into the real-world ways ADHD affects people's lives beyond just what medical websites say.
To answer your questions, though:
* Medication vs coffee: yes, I don't notice any effect from caffeine
* Meds weakening over time: nope
* Medication cost: so worth it (£45/mo for the drugs alone in the UK) because I was increasingly not able to adapt or cope and continuing to try to do so may well have destroyed me
I know I might not have ADHD, but I happen to be a magnet for people who do so it naturally peaks my curiosity as they are all considered to have ADHD, but have wildly different experience.
I don’t think they’re lying, but it is very clear that ADHD has entered the common vernacular and is now used as a generic term like OCD.
People will say “I’m OCD about…” as a way of saying they like to be organized or that they care about some detail.
Now it’s common to say “My ADHD made me…” to refer to getting distracted or following an impulse.
> or do you think it's possible that our pursuit of algorithmic consumption is actually rewiring our neural pathways into something that looks/behaves more like ADHD?
Focus is, and always has been, something that can be developed through practice. Ability to focus starts to decrease when you don’t practice it much.
The talk about “rewiring the brain” and blaming algorithms is getting too abstract, in my opinion. You’re just developing bad habits and not investing time and energy into maintaining the good habits.
If you choose to delete those apps from your phone or even just use your phone’s time limit features today, you could start reducing time spent on the bad habits. If you find something to replace it with like reading a book (ideally physical book to avoid distractions) or even just going outside for a 10 minute walk with your phone at home, I guarantee you’ll find that what you see as an adult-onset “ADHD” will start to diminish and you will begin returning to the focus you remember a decade ago.
Or you could continue scrolling phones and distractions, which will probably continue the decline.
This is a good place to note that a lot of people think getting a prescription will fix the problem, but a very common anecdote in these situations is that the stimulant without a concomitant habit change just made them hyperfocus on their distractions or even go deeper into more obsessive focus on distractions. Building the better habits is a prerequisite and you can’t shortcut out of it.
> Focus is, and always has been, something that can be developed through practice. Ability to focus starts to decrease when you don’t practice it much.
> The talk about “rewiring the brain” and blaming algorithms is getting too abstract, in my opinion. You’re just developing bad habits and not investing time and energy into maintaining the good habits.
> If you choose to delete those apps from your phone ...
I would like to add that focus is one of the many aspects of adhd, and for many people, isn't even the biggest thing.
For many people, it's about the continuous noise in their mind. Brown noise or music can partly help with parts of that.
For many, it's about emotional responses. It's the difference between hearing your boss criticise you and getting heart palpitations while mentally thinking "Shit, I'm going to get fired again", vs "Ahh next time I'll take care of this specific aspect". (Googling "RSD ADHD" will give more info.)
It's the difference between wanting to go to the loo because you haven't peed in 6 hours but you can't pull yourself off your chair, and... pulling yourself off your chair.
Focus is definitely one aspect. But between the task positive network, norepinephrine and the non-focus aspects of dopamine (including - more strength! Less slouching, believe it or not!), there are a lot of differences.
Medications can help with many of these, albeit at the "risk" of tolerance.
(I agree this is a lot of detail and nuance for a random comment online, but I just felt it had to be said. Btw - all those examples... might've been from personal experience - without vs with meds.)
this is an older thing than "I'm OCD when ..."
The AI models are just tools, but the providers who offer them are not just providing a tool.
This also means if you run the model locally, you're the one liable. I think this makes the most sense and is fairly simple to draw a line.
I had it stop right there, and asked it to tell me exactly where it got this information; the date, the title of the chat, the exact moment it took this data on as an attribute of mine. It was unable to specify any of it, aside from nine months previous. It continued to insist I had ADHD, and that I told it I did, but was unable to reference exactly when/where.
I asked “do you think it’s dangerous that you have assumed I have a medical / neurological condition for this long? What if you gave me incorrect advice based on this assumption?” to which it answered a paraphrased mea culpa, offered to forget the attribute, and moved the conversation on.
This is a class action waiting to happen.