So in a fucked up way, those TikTok kids swallowing Tide pods a few years ago are smarter than Gemini recommending “yummy button mushrooms”?

Damn that's a badass name tho... 🤘💀🤘

They can save people if caught in time. First things first, eat charcoal, crush and mix with water and drink. Not briquettes either actual charcoal it soaks up toxins and you excrete them instead of absorbing.

But the hospitals have liver protective substances, including milk thistle root extract they give.

iNaturalist is amazing (not necessarily their auto-id model, but the community that helps ID things).

I dream of the day we could put Nestle to death at the Hague

I don't believe in the death penalty, but I do believe in the corporate death penalty.

gives 3 million in small unmarked bills

'actually no wrongdoing was found'

Corporate personhood implies the opposite of what you want though. It means that the corporation can enter into contracts and be sued for breaches of them, but you want the officers of a corporation to be sued and held responsible directly.

Your business is directly, knowingly responsible for someone’s death? business license terminated

Even with corporate personhood, a business can't know things, only its employees can. If an employee is directly and knowingly responsible for someone's death, they should go to jail - but they already can be.

But if you're talking about the current example, who would be directly responsible for a death occurring because of a misidentified mushroom?

Hmm this person knew the victim, had a motive, and was really into mycology, and the autopsy had signs of poisoning from a rare mushroom…

Could always shoot a ricin dart from an umbrella gun. 🤷‍♂️

Here I made you some mushroom pasta!

It was wild reading that as it happened. "Oh no, only the people I invited were poisoned! I'm so lucky to be alive!" Homie. That's suspicious as hell.

Where is the style in that? Where is the panache?

Everybody is going to remember the mushroom murderer, nobody is going to remember some bleach boy.

Yeah, but with a mushroom there is a natural explanation or may not be as obvious as "someone fed this guy bleach".

People will eat a mushroom pizza if you give it to them

Well I'm already in a murderous mood, it's probably best I talk to as fewer people as possible.

That's a very metal name for a mushroom

Part of the problem here is that AI is mostly done by companies with billions of investments and in turn they NEEEEEDDDDD engagement, so they all made their AI as agreeable as possible just so people would like it and stay, with results like these becoming much more "normal" than it should or could be

I wonder how much of that is intentional vs a byproduct of their training pipeline. I didn't keep up with everything (and those companies became more and more secretive as time went on), but iirc for GPT 3.5 and 4 they used human judges to judge responses. Then they trained a judge model that learns to sort a list of possible answers to a question the same way the human judges would.

If that model learned that agreeing answers were on average more highly rated by the human judges, then that would be reflected in its orderings. This then makes the LLM more and more likely to go along with whatever the user throws at it as this training/fine-tuning goes on. Instead of the judges liking agreeing answers more on average, it could even be a training set balance issue, where there simply were more agreeing than disagreeing possible answers. A dataset imbalanced that way has a good chance of introducing a bias towards agreeing answers into the judge model. The judge model would then pass that bias onto the GPT model it is used for to train.

Pure speculation time: since ChatGPT often produces two answers and asks the user which one the user prefers, I can only assume that the user in that case is taking the mantle of those human judges. It's unsurprising that the average GenAI user prefers to be agreed with. So that's also a very plausible source for that bias.