Spyke

Replies

Comment on

Delta Will Use AI To Cut Costs And Set A Different Ticket Price For Every Passenger—CEO Says Profits Could Rise 50%

Reply in thread

You're getting downvoted, but you're right. This knife cuts both ways, and it far predates llms.

Any non trivial algorithm intended to rank is gameable.

This is the foundation of SEO. It's how the Donald invaded reddit. It's the fucking central pillar of modern propaganda dissemination.

They're going to spend a ton of money to lock themselves into an ongoing arms race.

And if one is morality opposed to this type of pricing practice, I think one becomes morality obligated to do exactly what you described.

And honestly, I think you don't even have to come out ahead for it to be worth it. If your LLM assessed price would be 1000, spending a full $200 on compute to game and $800 ticket is a wash. You SHOULD spend $300 to get an $800 ticket if that's where the floor is. Because fuck em.

Comment on

AI discovers that not every fingerprint is unique

Reply in thread

That is not at all what this article is about. The headline is terrible.

The research is suggesting that there may exist "per-person" fingerprint markers, whereas right now we only use "per-finger" markers. It's suggesting that they could look at two different fingers, (left index and right pinky, for example) and say "these two fingerprints are from the same person".

When they say "not unique", they mean "there appear to be markers common to all fingerprints of the same person"

Comment on

Avatar is cursed for remakes like Fantastic Four

Reply in thread

I think the main problem is that they keep inviting the original creators, they sign on, the studio heads explain to the creators how the studio has figured out how to tweak things, creators say "your ideas are horrible and if you execute on them as you've described, everyone is going to hate it". The studio's refuse to budge, creators depart citing creative differences, studio gets their way. Is a steaming pile of shit. Rinse and repeat.

Comment on

***hardcore leftist ideology***

I feel like this isn't super new. You haven't been able to do things like talk academically about network security or cryptography without being incredibly critical of government for decades.

My professors sure were. I felt really bad for them, honestly. They were the driving force to make philosophy and ethics courses mandatory at my university in order to recieve a comp sci degree.

"If you work in these areas long enough, at a high enough level, you WILL be asked to compromise something for the benefit of a state actor, and you might not even realize it's happening. The person asking you to do it might not even realize what is really happening." Was what our prof told a packed lecture hall.

Comment on

Over just a few months, ChatGPT went from correctly answering a simple math problem 98% of the time to just 2%, study finds. Researchers found wild fluctuations—called drift—in the technology’s abi...

Reply in thread

  1. It isn't and has never been a truth machine, and while it may have performed worse with the question "is 10777 prime" it may have performed better on "is 526713 prime"

ChatGPT generates responses that it believes would "look like" what a response "should look like" based on other things it has seen. People still very stubbornly refuse to accept that generating responses that "look appropriate" and "are right" are two completely different and unrelated things.