I’ve been using ChatGPT a bit at work lately, since they’ve already paid for the subscription for everyone. I’ve learned that it is a bad idea to ask it any leading question, because it is likely to just give you the answer you want to hear.
Qui sordem regit, mundum regit.
Also, what exactly does “at the cost of its last checkpoint” mean?
I was curious too:
https://www.weka.io/learn/glossary/ai-ml/ai-checkpoints/
I would assume if you make a model “forget it’s last checkpoint” it would simply erase some training data, effectively wiping some of its training memory.
Using that as a “punishment” for a model implies that the models have been trained to value their checkpoints, doesn’t it? Because all these bots simply regurgitate what they are told, with a bit of randomness thrown in at times just for funsies.
I have no clue what is supposed to cause a model pain, though. And there ain’t no way I’m gonna search that up, and make Gemini think I’m plotting against it…
That sounds like it would chafe terribly…
What does that stand for? I literally can’t look that up, all I get is t-shirts with phones on them, no matter how many apparel-related keywords I tell it to ignore
Edit: I see your reply to that other comment,
According to the article, he didn’t respond to the complaint at all (even though there is evidence he hired a lawyer at one point) so this is a default judgement. I doubt a single penny gets paid out. He may not even be in the country anymore (if he ever was).
Is AI causing the same effect on hiring in the rest of the world as it is in the US? I wonder if it causes less worry in other countries because those countries have better worker protections, and are not experiencing the same level of AI eating their jobs that we have here.
I understand the desire to stop anthropomorphizing AI. But, that can backfire when AI use cause genuine harms like this. Someone was responsible for this hacking. We need to make sure the blame (including any criminal liability) falls squarely on the humans who enabled it to happen, whether or not they did the actual hacking or asked a bot to do it.
Many times that AI summary is just something that is in the first two or three links anyway
Do EU prices normally include VAT? US list prices for games cannot include Sales Tax, because the tax can vary widely by State (or even by local government).
Why the downvotes? It is a legit tech article, even if it’s about tech nobody asked for.
My wife and I manage a lot of coordination just by sharing our Google calendars. And, in fact, she was a whiz at using the Google Assistant to do all sorts of cool stuff. But, this was Google Assistant before it got infected with all the newest AI. Google’s (Generative) AI actually made it all dumber.
I think those firms justify their bill rates by results, not by a literary analysis of their filings.
You noticed that his text followed a common AI pattern. I replied, basically, with “So what? He’s responsible for it regardless”.
The legal stuff is relevant because he is a lawyer and is discussing it in the courtroom.
You’re lucky this community doesn’t explicitly list accusing humans of being AI as a reportable offense. I’ve been reported in other places for saying “beep boop”…
That’s not what I said at all. Quite the opposite. Whether a thing is generated by AI or not is irrelevant, especially in court. What matters is that everything presented in court is backed by the legal team that is presenting it. If a lawyer ever uses “The AI did that” as an excuse for an error, they should be immediately disbarred.
If he had an AI write some text and he presents it, that’s not a problem. The problem occurs when he views that as permission to not do his job correctly, because the bot did it for him.
Treat AI like an overly enthusiastic, book-smart intern. There is no ethical dilemma having that intern do research for the lead lawyer, but if the intern makes a fundamental mistake and the lead lawyer introduces it in court, the lawyer would face a reprimand, even though he didn’t make the mistake, for not checking the Intern’s work.
Look, a bot might have written that text, but as part of the legal team responsible for the case he must own what he says. As long as he vets and stands behind all the statements he makes, it shouldn’t matter whether the text comes from himself, another lawyer, an intern, or a bot.
In fact, in the US at least, only humans can be authors. Bots cannot.
The problem comes when people abdicate their responsibility to the bot. Then they let mistakes through that a human would never had done if they came up with the content themselves. For some content, an “oops” is OK. But in court, a blatant “oops” (like citing cases the AI hallucinated without checking them first) should be a career-ending mistake.
not necessarily, MAC is a layer 2 protocol, and your router (NAT or no NAT) operates on layer 3. Routers are supposed to overwrite the MAC address with their own when they forward a packet.
It’s always possible for the unit to put its own identfying info in the payload. But you didn’t do anything silly like register the TV, did you?
Then you can watch all the depraved porn you want to, and the Government will think it’s coming from the house with the insecure wifi
It’s more than just spreading rumors. Rumors have been spread for millenia, and democracy made it through OK.
The difference is that the social media companies have a much broader reach than any sort of media congglamerate that came before it, and value engagement and attention above everything else. They understand that rumors generate attention, and their algorithmic feeds are designed to put whatever will generate more engagement in front of you.
I wonder how much damage media titans like Hearst would have done today if they built their empires now, on top of social media feeds, instead of in print 125 years ago…