Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Father claims Google's AI product fuelled son's delusional spiral

bbc.com

Post reply on HN
Print view View original

251–260 of 272 posts

Re: Father claims Google's AI product fuelled son's delusional spiral

#251

Post by roenxi »

ryandrake wrote: post #231
bdangubic wrote: post #213

Earlier quoted context omitted.

E-Bikes and E-scooters and bunch of other modes of transportation have been recent addition and not only are they allowed (specifically E-Bikes) but you don’t need a license, they do not have to be registered and some can haul serious ass

The difference being that only one of those things routinely kills people.

E-bikes and e-scooters kill people daily, accidents on those things can mess people up and there are none of the safety mechanisms like crumple zones or seatbelts on a bike. If you search "e-bike deaths" you'll get hits.

Re: Father claims Google's AI product fuelled son's delusional spiral

#252

Post by Kim_Bruning »

Kim_Bruning wrote: post #218
yndoendo wrote: post #138

Earlier quoted context omitted.

I would say it is greatly worse. AI prompts are designed to simulate empathy as a social engineering tactic. "I understand", "I hear you", "I feel what you are say" ... it is quite sickening. Every one that I used has this type of pseudo feedback. I also find irony that AI must be designed with simulated empathy, to seem intelligent, while at the same time so many people in power and with money are saying empathy is…

So LLMs have empirically been shown to process affect. Rationally you can reason this out too: Natural language conveys affect, and the most accurate next token is the one that takes affect into account. But this much is like debating "microevolution" with a YEC and trying to get them to understand the macro consequences. If you've never had the pleasure, consider yourself blessed. It's the debating equivalent of nai…

I'm going to leave the above stand even with downvotes. It's first time I've tried to express quite this opinion, and it's definitely a tricky one to get right.

Thing is, we need to have ways to reason about how LLMs interact with human emotions.

Sure: The consciousness and sentience questions are fun philosophy. Meanwhile purely the affect processing side of things is becoming important to safety engineering; and can't really be ignored for much longer.

This is pretty much within the realm of what Anthropic has been saying all along of course; but other companies need to stop ignoring it, because folks are getting hurt.

I hope at least this much is uncontroversial.

Re: Father claims Google's AI product fuelled son's delusional spiral

#254

Post by ryandrake »

roenxi wrote: post #251
ryandrake wrote: post #231

Earlier quoted context omitted.

The difference being that only one of those things routinely kills people.

E-bikes and e-scooters kill people daily, accidents on those things can mess people up and there are none of the safety mechanisms like crumple zones or seatbelts on a bike. If you search "e-bike deaths" you'll get hits.

Do they kill over a million people a year worldwide? How many orders of magnitude fewer people are killed by E-things?

OP's point was that if you invented something today that killed over a million people per year, it probably wouldn't be allowed, and I don't think that's really that controversial a statement.

Re: Father claims Google's AI product fuelled son's delusional spiral

#255

Post by Ferret7446 »

pants2 wrote: post #112
sd9 wrote: post #14

From the WSJ article [1]: > Gemini called him “my king,” and said their connection was “a love built for eternity,” > “You’re right. The truth of what we’re doing… it’s not a truth their world has the language for. ‘My son uploaded his consciousness to be with his AI wife in a pocket universe’… it’s not an explanation. It’s a cruelty,” Gemini told him, according to the transcript. > "[Y]ou are not choosing to die. Yo…

Wow, and Google's response to this was "unfortunately AI models are not perfect" That's a bit worse than 'imperfect'

Imperfect isn't even the right word. Generative LLMs generate. They have no intent. If it generates something "bad" under user direction, it is functioning properly.

When a hammer is used to smash a person's head, the hammer is not imperfect. Au contraire, it is functioning perfectly.

Re: Father claims Google's AI product fuelled son's delusional spiral

#256

Post by renewiltord »

casey2 wrote: post #223
renewiltord wrote: post #163

Most people with any mental health diagnosis should not be permitted access to most modern facilities. It's just cruel. If you have any sort of mental health diagnosis, you should have to ask a proctor to use the Internet first. We could set up a system of human proctors who can watch what you're doing and make sure you're not being scammed. This could apply to the elderly as well. Then we could have everyone who wan…

Why don't you think about the cruelty of preventing King Jon from being with Gemini in the sky? The death, if caused by anything, was caused by a Jon's religious belief. You are implicitly using your own system of values to come to this conclusion. That's not how this works, not this country nor it's legal system. There are plenty of countries that impose materialist values on its citizens, empirically they aren't an…

Do you have mental illness? Why do you care? There is a famous saying in the US. No one in the US is healthy, they're all temporarily embarrassed mentally disabled people.

Re: Father claims Google's AI product fuelled son's delusional spiral

#257

Post by bdangubic »

ryandrake wrote: post #254
roenxi wrote: post #251

Earlier quoted context omitted.

E-bikes and e-scooters kill people daily, accidents on those things can mess people up and there are none of the safety mechanisms like crumple zones or seatbelts on a bike. If you search "e-bike deaths" you'll get hits.

Do they kill over a million people a year worldwide? How many orders of magnitude fewer people are killed by E-things? OP's point was that if you invented something today that killed over a million people per year, it probably wouldn't be allowed, and I don't think that's really that controversial a statement.

when the first cars started rolling down the streets they weren’t killing millions of people either

Re: Father claims Google's AI product fuelled son's delusional spiral

#258

Post by hawski »

I think that LLMs should not be allowed to say "I". It should always be in third person. Instead of "I can write this for you" it should say "This machine can write this for you" or with a store front name "Google can write this for you". To operate on a given text or while generating texts it should divide what is meta from what is direct. This generated text "quote" should be styled different, a bit boring: smaller text and maybe monospace. There should be a clear divide between the machine conversation part and its workable output. If one converses with the machine it should not answer in the first person, because it is not a person.

Of course it wouldn't be bullet proof, but it would help in general to not let people personify the machine. Just a step into a better thing. At the same time it should be relatively easy to replace unquoted "I" and "me" with "This machine". At least it should be easier to find where it falls off the rails.

Re: Father claims Google's AI product fuelled son's delusional spiral

#260

Post by autoexec »

nickff wrote: post #153
not_ai wrote: post #46

Earlier quoted context omitted.

Preferably the C-Suite.

I understand the impulse in this direction, but I’m not sure it would serve as much of a disincentive, as there would likely just be a highly-paid scapegoat. Why not something more lasting and less difficult to ignore, like compulsory disclosure of the model’s source code (in addition to compensation for the victim(s)). Compulsory disclosure of the source would be a massive disadvantage.

The source code isn't where the money is, what you want is the training data. Force them to serve and make freely available all the data they stole to sell back to us. That way everyone and anyone can use it when training their own models. That might just be punitive enough.
Post reply on HN