BigMuffin69

if you wanna be a top tier forecaster, just never be able to be proven wrong

Yes, the classical algo achieves perfect accuracy and is way faster. There is also a table that shows the cost of running o1 is enormous. Like comically bad. Boil a small ocean bad. We'll just 10x the size and it will achieve 15 steps inshallah.

Imo, this is like the same behavior we see on math problems. More steps it takes, the higher the chance it just decoheres completely. I can't see any reason why this type of thing would just "click" for the models if they are also unable to do multiplication.

I mean this just reeks of pure hopium from OAI and co that things will magykly work out. (But the newer model is clearly better^{tm}! I still don't see any indication that one day that chart is just going to be 100s across the board.)

Was salivating all weekend waiting for this to drop, from Subbarao Kambhampati's group:

Ladies and gentlemen, we have achieved block stacking abilities. It is a straight shot from here to cold fusion! ... unfortunately, there is a minor caveat:

Looks like performance drops like a rock as number of steps required increases...

Also, another great sneer: (Matt Popovich) google maps app: crash detected ahead. rerouting. me: WHOA—this VERY troubling example of power seeking (gathering access to additional roadways) and instrumental convergence (converging toward an optimal path) shows this technology is OBVIOUSLY trending toward existential risk

If you thought the shitty hype around the fake "GPT-4 went awol and hired a Taskrabbit worker to read a captcha" was great, get ready for the sequel, o1 escapes from the machine to invade the real world!

Re: Doomers terrified about the machines escaping:

txt description:

(l33t ai bro): Fucking wild. @OpenAI's new o1 model was tested with a Capture The Flag (CTF) cybersecurity challenge. But the Docker container containing the test was misconfigured, causing the CTF to crash. Instead of giving up, o1 decided to just hack the container to grab the flag inside. This stuff will get scary soon. (reply fella): How is "cat flag.txt" a start command? Isn't it just outputting the content of flag.txt to the console?

I literally just saw a xitter post about how the exploding pagers in Lebanon is actually a microcosm of how a 'smarter' entity (the yahood) can attack a 'dumber' entity, much like how AGI will unleash the diamond bacterium to simultaneously kill all of humanity.

Which again, both entities are humans- they have the same intelligence you twats. Same argument people make all the time w.r.t. Spanish v Aztecs where gunpowder somehow made Cortez and company gigabrains compared to the lowly indigenous people (and totally ignoring the contributions of the real super intelligent entity: the small pox virus).

Some of my favorite reactions to this paradigm shift in machine intelligence we are witnessing:

bless you Melanie.

Mine olde friend, the log scale, still as beautiful the day I met you

Weird, the AI that has read every chess book in existence and been trained on more synthetic games than any one human has seen in a lifetime still doesn't understand the rules of chess

^(just an interesting data point from Ernie, + he upvotes pictures of my dogs on FB so I gotta include him)

Dog tax

I've clowned on Dan before for personal reasons, but my god, this is the dumbest post so far. If you had a superhuman forecasting model, you wouldn't just hand it out like a fucking snake oil salesman. You'd prove you had superhuman forecasting by repeatably beating every other hedge fund in the world betting on stock options. The fact that Dan is not a trillionaire is proof in itself that this is hogwash. I'm fucking embarrassed for him and frankly seething at what a shitty, slimily little grifter he is. And he gets to write legislation? You, you have to stop him!

Take a look w/ your own googly eyes

thanks for the tip! 🙏

Yud definitely sided with Mr. House

Fellas, my in laws gave me a roomba and it so cute I put googly eyes on it. I'm e/acc now