_mitterpach

Umm, no? There is no love to be gained, there is no attraction. This is only a dangerous trap that offers lonely and unloved people a simulation of what it might be like to be in a long-term commited relationships.

I remember reading a Reddit thread about a woman wanting to date ChatGPT, because compared to her boyfriend, GPT was talkative, supportive and telling her all the right things. Why would you continue to date somebody who does not support you in the ways you want?

If you're using these applications to simulate anything regarding human social connections, please shut your computers down immediately. Find a hobby, join your nearest D&D club, just please do not become emotionally dependent on a simulation of a person, provided to you by complete leeches, who will only try to milk this connection for more and more money.

[Ben Buchanan]: A canonical definition of A.G.I. is a system capable of doing almost any cognitive task a human can do. I don’t know that we’ll quite see that in the next four years or so, but I do think we’ll see something like that, where the breadth of the system is remarkable but also its depth, its capacity to, in some cases, exceed human capabilities, regardless of the cognitive discipline —

[Ezra Klein]: Systems that can replace human beings in cognitively demanding jobs.

[Steve]: Yes, or key parts of cognitive jobs. Yes.

The interview was really interesting, and it's very revealing to see what's in the head of somebody advising on the future policy. That said, most policy-making actors have some kind of stake in the results. That should probably be kept in mind when looking at the narrative that's coming out.

That said, this is not software engineering. It's tinkering, it's fun, and it can get you a foot in the door when it comes to approaching real work, but it's no way to build a business. As soon as the context window is not enough, you run into so many problems everything comes to a standstill. I don't think the article makes this distinction in any meaningful way, and the narrative is very quickly forming to say that we simply don't need any coders, any junior developers. This is false, and furthermore, this is dangerous.

Can you replace a junior with LLM? Probably, juniors are often a drag and a net minus. Should you do that? Absolutely not! You need midlevels, you need seniors, and both of those do not just magically appear. They need to be grown, taught, nurtured. We'll see the ripples of today's shortsighted thinking eventually.

Could this have been one of the first signs of misleading, lies and fraud?

Do you think Anthropic is still a startup? That's an honest question, Wikipedia certainly seems to think so, (Anthropic PBC is an American artificial intelligence public-benefit startup founded in 2021.). I would typically say that once you have 61B valuation, you have to have a fundamentally sound business, and thus cease to be a startup, but I don't know where the line is drawn.

But how do you decide who votes? The fairest is obviously to allow voting for everyone. As soon as you try to single out 'unsuitable' people, you run into a problems of how you would do so? You could try to give people a basic test, but these 'literacy' tests were historically used to supress all kinds of minorities.

That is the true problem of democracy, it sucks, but everything else is miles worse. The people voted, the people get what they voted for. If you want a different result, convince people to vote differently.

This entire saga is mind-blowing to watch. I firmly believe LaLiga is completely in the wrong here and are trying to scare pirating customers into paying by suing everyone even remotely connected to this. Their behavior actually makes the internet worse for their paying customers as well.

Any remotely competent judge would shut this behavior down immediately, but money talks in this case I guess. I'm interested in seeing how far they get in these court cases. Could anybody knowledgable in Spanish law take a guess as to how these cases will end?

I don't like having such a black and white view of the world though, so I would be interested in seeing if there is any sane reason for these valuations. Do you think these businesses would be as highly-valued if there was not all this hype around? How would you justify this valuation?

The way I see it, either one of two options will happen:

1. Trust based assessments Everything will be continue as is now, only the students will be asked nicely not to use any AI in their learning, with the explanation that they're only shooting themselves in the foot by not learning anything that will be required later in life.

2. Back to basics Everything will be paper-based again, with only in-person tests being used for grading. This sucks for a lot of reasons.

I don't see the education field innovating in any way, especially if it means adding more work on top for teachers, in the form of designing tests specifically to be AI-proof. The entire educational system stands on the shoulders of low-paid, overworked teachers, who are burnt out already.

It feels like Europe fell asleep when it comes to space innovation, but at this point reusable launches are not even an innovation, they're just how the business is done. And until we manage to build reusable infrastructure and rockets, the costs are going to be astronomical compared to the competitors.

Perhaps with the situation now, Europe could snag some rockstars from US?

Has anybody actually found a use for tech such as Apple Intelligence? What could actually Siri do that would make it worth using?

The article does not talk about the actual state of lab-made viruses. Could anybody tell me (in the least politically influenced terms), what is the current state of man made viruses, is it feasible to create Covid-25 or anything similar for a human? What would be needed, a large multinational laboratory complex, or just a small homemade device?

But this does point to a weird trend in the industry, where the bar is being artifically raised by people who attempt to cheat at interviews. At my job we had one interview where the guy was visibly reading off of ChatGPT (reflection in glasses, weird stops).

To HN: What alternatives are there for estimating interviewee's skillset, what's your favorite one?

I can’t imagine the same thing happening in any other industry. Your company asking you to sign away code that you’ve written, so an AI can be trained on it?

These companies are missing the whole point, and a VA’s job is a lot more than just saying the words. Just look at how the cast of Baldur’s Gate is still doing promotions from the game, one year from launch. People love it, and industry could stand to learn a lot from BG3’s success.

https://gitlab.com/librespacefoundation/satnogs

There's over 600 open issues with everything from hardware to website UX, so depending on your field there may be a lot you can contribute.

I am a huge fan of everything coming out of China and Europe lately. OpenAI needed competitors, this forces them to actually deliver graspable bussiness benefits, not just smoke and mirrors.

Hope this one goes differently to my landings in Kerbal Space Program.

From the wiki I read that this is a part of the Artemis program, but the connection seems unclear to me. Could anybody explain, will this be used to deliver just cargo or will it have people aboard sometime in the future?

We’re doing something similar, as it is by miles ahead of any other options. I’m looking forwards to seeing the landing live.

It is outstanding how much soft power have both US and Russia lost in this decade. Russia used to be a feared military state, now they cannot take Ukraine in three years and their army is made of paper. US has abandoned its allies and is rotting from the inside.

China? They have been investing their money all over the world, Africa, South America, Europe. They seem like they are the only ones who remember how the diplomacy game is played, meanwhile Europe can’t make decisions and focuses on the most meandering matters.

Then there’s everything else, like not having an offline mode, taking a long time to load and everything else. Anecdotally, I know maybe a dozen people who gave Notion a try and then just.. stopped using it.

Is anyone there actually still using Notion? Have they ironed out the issues or have they just added a bunch of AI stuff that you immediately turn off?

You mention briefly displaying metrics on office dashboard TV screens, that's a really fun idea. Is it just a bunch of Grafana graphs, or is the system more in depth? I might have to think about creating something similar at my place.

Obviously the link is primarily talking about the fact that the player function is 4000 lines long. That's a lot. Maybe that's the lesson, your code does not have to be perfect, the result must.

I'll also add to the discussion this amazing video by GMTK, which explains what video game developers must think about when designing movement. It's wonderful.

Why Does Celeste Feel So Good to Play? (https://www.youtube.com/watch?v=yorTG9at90g)

For a video that's (surprisingly) closely related to the topic, see CGP Grey's video on the flag of Antarctica:

https://www.youtube.com/watch?v=U0wTDK0VOeY

(The most common flag used for Antarctica was created for the 3D World Atlas software, so presumably whoever owns the copyright to that software now, owns the copyright to the flag of Antarctica)

That said, what is the actual crime they are going to be charged with? There are no actual laws preventing this, the article even states:

Operation Cumberland has been one of the first cases involving AI-generated child sexual abuse material (CSAM), making it exceptionally challenging for investigators, especially due to the lack of national legislation addressing these crimes.

And even if it was, surely you must look at the model makers themselves as well. Further on, the article even states the following:

Even in cases when the content is fully artificial and there is no real victim depicted, such as Operation Cumberland, AI-generated CSAM still contributes to the objectification and sexualisation of children.

In the most honest manner possible, is there anyone that could explain what factual laws are broken here? There may even be an argument to be made that this is better because no actual children are harmed?

Note that this only exposes the repositories that were public at one point, then made private. While they were public, the contents got indexed and the ever-data-hungry models can still access this indexation.

We informed Microsoft of our findings, outlining that deleted or private GitHub repositories were still accessible through Bing’s cache and Copilot and that this functionality posed a security risk by exposing organizational data. We also alerted all affected organizations and advised them to rotate or revoke any compromised keys.

Microsoft classified the issue as low severity, citing its "low impact" and asserting that this caching behavior was acceptable. Nevertheless, they responded quickly and within a two weeks, Bing’s cached link feature was removed, and access to the cc.bingj.com domain was disabled for all users.

From the article, I'd probably disagree on the low impact, but nontheless the response is reasonable.

Can anybody who's looked into this topic further tell me why this may not be the case? Would the pressure down there be too big, would it be too cold or what other possibilities could prevent a planet with oceans to manifest life?

Sure, there are some coding competitions, there are some benchmarks, but can you really check if the recipe for banana bread output by Claude is better than the one output by ChatGPT?

Is there any reasonable way to compare outputs of fuzzy algorithms anyways? It is still an algorithm under the hood, with defined inputs, calculations and outputs, right? (just with a little bit of randomness defined by a random seed)

“We’ve been growing a lot and are out of GPUs,” Altman wrote. “We will add tens of thousands of GPUs next week and roll it out to the Plus tier then … This isn’t how we want to operate, but it’s hard to perfectly predict growth surges that lead to GPU shortages.”

So they are 'out of GPUs', for this week only? Why would they push out the launch then? Could anybody provide more insight?

Is this only for those that are fully self-driving?