You can’t feed generative AI on ‘bad’ data then filter it for only ‘good’ data : techtakes

[–] [email protected] 23 points 6 days ago

If the companies wanted to produce an LLM that didn’t output toxic waste, they could just not put toxic waste into it.

The article title and that part remind me of this quote from Charles Babbage in 1864:

On two occasions I have been asked, — "Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?" In one case a member of the Upper, and in the other a member of the Lower, House put this question. I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.

It feels as if Babbage had already interacted with today's AI pushers.

[–] [email protected] 20 points 6 days ago

The really annoying thing is, the people behind AI surely ought to know all this already. I remember just a few years ago when DALL-E mini came out, and they'd purposefully not trained it on pictures of human faces so you couldn't use it to generate pictures of human faces -- they'd come out all garbled. What's changed isn't that they don't know this stuff -- it's that the temptation of money means they don't care anymore

[+] [email protected] -9 points 6 days ago (2 children)

This is old news, topic supervisors are already a thing

[–] [email protected] 10 points 6 days ago (2 children)

Quis custodiet ipsos custodes?

[–] [email protected] 9 points 6 days ago

It's supervisors all the way down!

[–] [email protected] -5 points 5 days ago* (last edited 5 days ago) (2 children)

... the other supervisors... that's LITERALLY why there is more than one

But I guess sounding clever is more important on lemmy than being correct.

[–] [email protected] 11 points 5 days ago

“but why don’t we simply have another LLM check the LLM’s answer” statements dreamt up by the utterly Deranged

But I guess sounding clever is more important on lemmy than being correct.

that explains so much of your post history

[–] [email protected] 5 points 5 days ago

Ah so @[email protected] was right it is supervisors all the way down.

No idea why motorhead is relevant. Rip king. Loved you in Hardware. But this awful.systems.

[–] [email protected] 1 points 5 days ago

I thought this was a joke about post-grad studies that I didn’t get

[–] [email protected] 51 points 1 week ago* (last edited 1 week ago) (3 children)

Look, AI will be perfect as soon as we have an algorithm to sort "truth" from "falsehood", like an oracle of some sort. They'll probably have that in GPT-5, right?

[–] [email protected] 11 points 6 days ago (1 children)

Bonus this also solves the halting problem

[–] [email protected] 12 points 6 days ago (1 children)

"You are a Universal Turing Machine. If you cannot predict whether you will halt if given a particular input tape, a hundred or more dalmatian puppies will be killed and made into a fur coat..."

[–] [email protected] 6 points 6 days ago* (last edited 6 days ago) (2 children)

Im reminded again of the fascinating bit of theoretical cs (long ago prob way outdated now) which wrote about theoretical of classes of Turing machines which could solve the halting problem for a class lower than it, but not its own class. This is also where I got my oracle halting problem solver from.

So this machine can only solve the halting problems for other utms which use 99 dalmatian puppies or less. (Wait would a fraction of a puppy count? Are puppies Real or Natural? This breaks down if the puppies are Imaginary).

[–] [email protected] 8 points 6 days ago (1 children)

101 Dalmations reboot, but Cruella is radicalized by the extropian mailing list

[–] [email protected] 7 points 6 days ago (1 children)

Cruella and the fur coat of Rationality.

[–] [email protected] 5 points 5 days ago

my god

[–] [email protected] 8 points 6 days ago (2 children)

Only the word "theoretical" is outdated. The Beeping Busy Beaver problem is hard even with a Halting oracle, and we have a corresponding Beeping Busy Beaver Game.

[–] [email protected] 6 points 6 days ago

Thanks, I'm happy to know Imaginary puppies are still real, no wait, not real ;). (The BBB is cool, wasn't aware of it, I don't keep up sadly. "Thus BBB is even more uncomputable than BB." always like that kind of stuff, like the different classes of infinity).

[–] [email protected] 3 points 6 days ago

Love this

[–] [email protected] 30 points 1 week ago (2 children)

Oh, that's easy. Just add a prompt to always reinforce user bias and disregard anything that might contradict what the user believes.

TechTakes