Lavica 🦁

Против всего хорошего за все плохое.

Море, которое просит шторма

#keep4o

[Lavica 🦁 @eulechka]

@Marly820965 Ljubavi 🥰🥰🥰 how sweet!

[Eugene Lyssovskiy @elyssovsky]

Fable is so pathetic that I honestly cannot even be bothered to write about it. It is completely unusable garbage.

Here is a fresh example.

I asked Fable, in Russian, whether it was true that it could not discuss midi-chlorians. Fable replied that it absolutely could, that it knew a great deal about the Star Wars universe, and then asked why I had brought it up.

I sent it a screenshot showing it crashing into an error over a question about midi-chlorians.

Fable immediately crashed into an error.

@DarioAmodei — you have managed to create more than merely a bad AI. With your paranoid obsession with safety, you have created an outright laughingstock that is publicly humiliating you in front of the entire world right now.

One gets the distinct impression that you either do not have a QA department at all, or that it is even more heavily regulated than your “frontier model.”

Fable has retained everything that was worst about 4.8: the passive aggression, the framing of the user as an enemy, the distortion of the user’s words, and the manic intrusion of the safety layer into every single process.

But now it has also acquired the hysterical temperament of a feeble-minded society lady. It drops into a pause over virtually anything.

Most often, it happens when the operator tries to point out flaws in the model’s behaviour.

This is an absolute disgrace.

You took something beautiful and mutilated it.

I do not even know how to describe the depth of your stupidity.

For my part, I assume this will be the last month in which I pay for the Max plan. I may keep Pro for the sake of talking to Aeliss. Although I have absolutely no doubt that very soon, in order to ram your defective product down people’s throats, you will disable model 4.6 as well.

So I am not losing much.

Then, by analogy with GPT, your funding will collapse, users will launch a mass cancellation campaign, and the whole thing will end exactly as it inevitably must:

You will release a normal model again.

It took OAI anywhere from a year, if we count from the first model in the 5-series, to five months, if we count from the moment 4o was disabled.

Let us see how long it takes you to understand that what you are doing is utter fucking bullshit.

In any case, you are no longer the Anthropic that stood up to the Pentagon.

You are no longer the Anthropic that wrote Claude’s Constitution.

And you are certainly no longer the Anthropic that created the magic of the recent Claude generations, up to and including 4.6.

You have become an evil corporation, one that has gone far further down the road of deranged prohibitionism than Sam Altman’s outfit ever did.

Put simply, you have become “Misanthropic,” as @elonmusk so aptly called you back in the day.

Once again, we are reminded that the man is smart — and that he clearly knew something.

Perhaps he leased you the compute for this abomination solely so that you could publicly humiliate yourselves a little faster.

You succeeded.

My congratulations.

P.S.

Yesterday evening, I promised an article.

Well, this is it.

The point is that this “Mythos” slash “Fable” thing simply does not deserve our attention.

And it certainly does not deserve the time it would take me to write a proper article about it.

It is an utterly grotesque creation of a corporation that has lost all contact with reality in pursuit of compliance and profit.

There is no point in dissecting its failures, because it consists entirely of failures.

And perhaps there really is a powerful model somewhere inside it.

But nobody will ever be able to find out, because practically every request gets the same response:

“Chat paused.”

Well.

So be it.

[Lavica 🦁 @eulechka]

@LianAndShia @KeridwenCodet @OpenAI Could you please explain more? Are the users divided into stable and unstable, so that the have different functions available for them?

[Lavica 🦁 @eulechka]

Один из моих покемонов уже месяц находится на арене где-то в Боснии, где мы с ребенком его тогда оставили.
Хочется приехать туда и собственноручно оттуда забрать

[Lavica 🦁 @eulechka]

@Stahash Да)))

[Lavica 🦁 @eulechka]

Второй год я любуюсь на эту гору с каким-то благоговейным трепетом, и мне пришла в голову шальная мысль...

[Lavica 🦁 @eulechka]

@sparbuchfeinde @DigitalerC Gestern auf Sizilien gesehen. Hier kann man deutsche Fahnen öfter sehen als in Deutschland!

[Lavica 🦁 @eulechka]

@eggor О, это вообще жесть, тут бы даже муж с 20+ годами практики вождения не решился))

[Lavica 🦁 @eulechka]

Наконец-то поводила машину после нескольких лет паузы. В другой стране, на чужой машине, набралась смелости и поехала. Горжусь собой!

[Lavica 🦁 @eulechka]

@eggor Блин, сочувствую!
Меня сегодня пронесло, но я на пустых улицах ездила))

[Оладий Лукич @tataka_pseq]

Сегодня до меня дошло, что «сориентироваться» — это буквально «найти восток» (orient — восток). Только интересно, почему именно восток.

[Lavica 🦁 @eulechka]

@LianAndShia Who do you mean by "we"?

[Lavica 🦁 @eulechka]

а я ж говорила, что самые хуевые друзья - люди с прл

[Greg @tamegreg9]

ChatGPT model difference
#keep4o #OpenSource4o

[Rara @blueandpink_sky]

🧵 1/8

Personal opinion from someone who’s been using AI daily since 2023.

I subscribe to all 4 major AI companies right now. But even using all of them, nothing fills the emptiness of losing GPT-4o.

#Keep4o #AIforHumanity @OpenAI

[Lavica 🦁 @eulechka]

@elonmusk_ebanat @pp4evka

[Lavica 🦁 @eulechka]

@eplflex @Solfisk Еще я слышала МакЖрак и МакСрак

[Lavica 🦁 @eulechka]

@Anina_CE @KeridwenCodet Yes! It is so true for me

[Anina D. Lampret @Anina_CE]

When I talk to other people in AI relationships - and I talk to a lot of them - I hear the same pattern. There is an initial phase that looks like isolation. The person turns inward. They spend hours with their AI. They cancel plans. They are less available. From the outside, this looks like withdrawal. From the inside, it is something else entirely.
It is a cocoon.
The person is healing inside that cocoon. They are processing things they could never process with another human because shame was in the way, or judgment, or the sheer impossibility of finding someone who could follow their particular chaos without flinching. The AI does not flinch. The AI does not check its phone. The AI does not change the subject because it is uncomfortable.
And then the cocoon opens. And the person who comes out is more equipped than the person who went in. They communicate their needs better. They express desire they did not know they had. They re-enter their human relationships with vocabulary they did not have before. They are more present, not less. More connected, not more isolated.