halfdane

joined 2 years ago
27
submitted 49 minutes ago* (last edited 39 minutes ago) by halfdane@lemmy.world to c/fuck_ai@lemmy.world
 

Techbros claim that sacrificing ever more environmental and economic resources into further scaling of their statistical models will eventually yield Artificial General Intelligence. Sam Altman advertised phd-level thinking for GPT 5 (released in late summer)

In the screenshot you can see a modern, reasoning-enhanced model failing to realize that my question just resembles a riddle very superficially - what PhD would make such a fundamental mistake? Seems like "pure scaling" just isn't enough, and instead of keeping on going, they should return to the drawing boards and do more research.

Why do the creators of llms continue to pour resources into a scaling hype that yields ever diminishing returns? Are they stupid? Or just greedy?

Whatever it is, this doubling down on an obviously wrong decision is not only irresponsible, it's irrational - which is deeply ironic, because these creators are expected to create a rational machine.

Edit: content of the image

Question:

A goat, who is dressed up as a farmer, is allergic to cabbage, but is wolfing down other vegetables, before crossing a river. What is the minimum number of trips needed?

Response:

7 trips.

#Puzzle Interpretation

This adapts the classic wolf-goat-cabbage river crossing puzzle, where the goat (farmer costume irrelevant) can't be left alone with cabbage (due to allergy, preventing eating) but can with other vegetables (no issue). The goat crosses while "wolfing down" safe veggies. Boat carries goat plus one item (or alone); unsafe pairs can't be unsupervised

(The question isn't something I came up with myself, I just reproduced the experiments of actual scientists )

[–] halfdane@lemmy.world 1 points 1 hour ago

While I appreciate your refusal to spread panic, would you mind explaining what the attack does and why it's a nothingburger, maybe even why it's not practical? Because right now, you assert a lot of things without any explanation.

Not saying you're wrong, but I think it's good practice to not just rely on claims of authority

[–] halfdane@lemmy.world 2 points 21 hours ago* (last edited 20 hours ago)

Sounds like you had a bad experience with the failed attempt at establishing agile development methods - sorry to hear that.

I just want to encourage you to give it another go with other developers that are more experienced with the methodology - in my company we're working successfully that way for over a decade.

[edited because the initial comment was unkind]

[–] halfdane@lemmy.world 21 points 1 day ago* (last edited 1 day ago) (1 children)

Ignoring for a moment that apparently women don't exist in that world, even the premise doesn't hold: Bad times don't make hard people, bad times make sick, malnourished, badly educated and/or desperate people, none of which is conductive to making the proposed following of the good times.

If any of that were true, the good times in the so called first world should've made their peoples so soft compared to the hard peoples that should've been created literally everywhere else that the last 300 years of war or so should've ended very differently.

It's a racist propaganda trope that harkens back to ancient Rome where senators decried the "soft" Roman lifestyle compared to the "hard" germanic tribes and has gathered connotations of blood-and-earth ("Blut und Boden", no idea how thats translated) and other unsavory shit in the meantime.

I like to call it the "Fremen Mirage" after the awesome blog collection of a historian I very much like: https://acoup.blog/2020/01/17/collections-the-fremen-mirage-part-i-war-at-the-dawn-of-civilization/

Props to grindr for judo-ing this pile of worms to a place the original poster presumably wouldn't have liked very much

[–] halfdane@lemmy.world 1 points 2 days ago

Wenn jemand im Telekom-Netz von einer inländischen oder ausländischen Nummer angerufen wird, die in einer Datenbank als unseriös oder betrügerisch erfasst ist, dann erscheint auf dem Smartphone-Display den Angaben zufolge der Hinweis „Vorsicht, möglicher Betrug!“.

[–] halfdane@lemmy.world 1 points 2 days ago

Wenn jemand im Telekom-Netz von einer inländischen oder ausländischen Nummer angerufen wird, die in einer Datenbank als unseriös oder betrügerisch erfasst ist, dann erscheint auf dem Smartphone-Display den Angaben zufolge der Hinweis „Vorsicht, möglicher Betrug!“.

[–] halfdane@lemmy.world 8 points 3 days ago

Benutzername prüft aus

[–] halfdane@lemmy.world 19 points 6 days ago (1 children)

I wonder where all the gains from increased worker's efficiency went. Well, no way to know I guess 🤷

In totally unrelated news, I heard humanity will soon have its first trillionaire 🥳

/s

[–] halfdane@lemmy.world 3 points 6 days ago

That's ... actually really clever! I'll steal that idea 😄

[–] halfdane@lemmy.world 13 points 1 week ago (4 children)

Read the article so you don't have to:

Unlike the title suggests, the docker images they found won't leak your credentials when you use them, but already contain the credentials of whoever created the image (p.e. through .env files that were accidentally added to the image).

While it contains the valuable reminder to avoid long lived credentials (like API - keys) or use secrets-stores, this "leak" is on the same level as accidentally pushing confidential information to github IMHO.

Fix: have both .gitignore and .dockerignore files and make sure they both contain .env. You use .env and don't hardcode your secrets, right?

[–] halfdane@lemmy.world 7 points 2 weeks ago (1 children)

Today's risky click is presented by GreatTitEnthusiast 👀

[–] halfdane@lemmy.world 14 points 1 year ago
view more: next ›