If this doesn’t mean graphics cards will get cheaper or more expensive I can’t say I care

I guess disclosing that a blog post was substantially written by AI is better than not disclosing it, but it sure is a choice to do that in the last paragraph

I’d object to this if it were open source and forkable, but looks like it is not and so your OS becoming corpo-fied is an especially salient and unavoidable risk:

Cortex is KORA’s own technology and is not open source

I think the top hackernews comment is probably right and this post itself is AI engagement farming, the user has their account set to hidden, it’s very pointedly feeding into a narrative and telling people what they want to hear, and the writing style is AI-ish even if it doesn’t have the more obvious tells.

Isn’t it enabled by default, and requires extra effort to set up key-based auth? Surely many people take the path of least resistance.

Ah, fair point, didn’t notice the suspicious qualities of the rest of the article

I feel like they explicitly prompted it to give that guy a fucked up arm, which is a nice touch if so

Customer-service tools, technical or research systems, and certain internal corporate uses are exempt under Senate Bill S9051B.

Does technical systems mean API based inference providers are off the hook or not? Otherwise I guess it’s time to break out the jailbreak prompts for the Chipotle chatbot.

Local law enforcement were made aware of the situation when West’s heirs tried to sell the property, according a police affidavit.

Really bizarre situation, not sure why he would have an expectation of getting away with that

The type of bias described in the article (more code vulnerabilities if the prompt mentions the code is for groups not aligned with the Chinese government) does not seem like it should be so complicated to avoid.

Advancing a favorable narrative does seem like the reason for this report, but knowing that Anthropic is very biased isn’t enough by itself to say their claims are bullshit. It’s still worth considering and critiquing the argument on its own merits, stopped clock is right twice a day and all that.

It’s not like there is a shortage of games on Steam that are willing to disclose that AI was used in their development and have already done so. The prospect of not disclosing and keeping that a secret seems like a really uncomfortable way to do things.

This is why things like EU kids act are targeting videogames

It also means that his praise for and indeed positive experience with this “tool” is largely irrelevant to the broader social/economic discourse surrounding so called AI.

Yeah, but tbf all of the statements he’s making about it are in the context of organizing linux development and the most effective way to carry that out, and just happen to be public and so end up being widely republished and incorporated into discussions with a broader topic than that.

That does change things a bit

I don’t see how this means he’s wrong though, if people are hating AI as a proxy for hating their abusive relationship with their employer, that doesn’t seem like a strong argument that everyone who isn’t in that situation must also hate AI.

Companies want you to associate your transactions with a phone number so they have data on who is buying what. For instance if you go to Safeway it’s common that the price of ice cream will be twice as expensive if you haven’t registered a phone number with them. But you can give someone else’s phone number that has already been registered and that will also work, or register a fake number, a number lots of people would have thought of is likely to have already been registered. At least that was how it worked last time I checked, that was a long time ago.

And if the traffic is closely monitored, you’d think they’d also notice all their cpu and ram suddenly being used for some mysterious purpose

Seems kind of impractical, why even bother with trying to propagate a 4b model running at 0.39 tokens per second if you could instead make a conventional worm using a more powerful AI model targeting known devices? It seems like the only reason you would want to do this is if you can only find vulnerabilities while actually on the given network, but that would be kind of niche since how many people are really running totally custom stuff?

Price-suppression, which includes cuts to tariffs and indirect taxes as well as price controls, can reduce inflation expectations “slightly”, the IMF authors concede. But these measures have little else to recommend them. They benefit the rich more than the poor for the simple reason that the rich buy more of the stuff the policies keep cheap. The IMF authors calculate that for every euro spent suppressing the price of electricity, gas and petrol in European countries, only 20 cents reached the poor. Price-suppression sometimes takes the form of subsidies to producers. These make downstream firms more competitive overseas, increasing their exports. The money, in other words, flows from the taxpayer to foreign customers. Even price controls, which do not require direct payments from government coffers, can impose fiscal costs in the long run. The state must often bail out utilities that cannot pass on the extra cost of their inputs.

Makes some sense but I feel like they are leaving out options like breaking up monopolies, or subsidizing the production of things like housing that can’t be exported, or directly producing various infrastructure (trains, renewables) that reduces the need for fossil fuels.