Factbook, not Facebook.

Does this indicate that they’re training it on the users’ words, and they don’t want it to learn bad habits?

Apparently giving anything to anyone but Trump is a snub to Trump.

The article is from September 17th. Who knows what happens to this policy now. But it’s still quite possible the USA will sanction other countries for buying Russian oil, because they want the oil for themselves.

He doesn’t intend to loosen his grip on power before he dies.

Wow, that is a cowardly take.

It’s the fascist slide: every day a new outrage, then by the next day it’s normalized. Everyone thinking “surely this time someone will do something to stop them.”

What a move, asking someone a question, then cutting them off every time they begin to answer with “Why won’t you answer the question?”

That license suggests these people don’t know what they’re doing, or don’t care. Maybe they think they’re geographically out of reach of Adobe’s lawyers. In any case, it’s hardly striking a blow for open-source software when its provenance and legal status are so dubious. FOSS should be done honestly and with pride.

You could argue that “vibe coded” and “clean room reimplementation” are by nature incompatible with one another, since you can’t prove the AI didn’t ingest the original code. But you could also question whether the AI could possibly be trained on Adobe’s code when Adobe keeps that under wraps. It could be trained on leaked code, but that would be hard to prove.

If they used decompilation as a guide for the AI, that might be easier to establish regardless of what the LLM was trained on. There could be telltale implementation details that Adobe could cite.

In any case, Adobe has a huge number of patents on its software, so I imagine there could be grounds for legal action just because these applications faithfully copy the look and feel and functionality of Adobe products. That might be easier for them to prove in court than anything about the code’s provenance.

It may be useful for the very specific kind of use case you describe, but that doesn’t help the AI companies with their claim that this needs to be a multi-trillion-dollar industry. I too have found it’s good for little applications I never intend to distribute, but as soon as I think of sharing a project, it’s clear that there’s quite a bit of slow work needed to check its quality and security.

That’s kind of the whole problem with AI in every domain: it can save you time if you let it churn out unchecked garbage, but if you check its outputs it won’t save you time and you’d have been better doing the work yourself. It’s still not clear whether there’s any room for actual value in between these.

Too many people (including my boss) seem to think AI is a magic oracle that can replace the whole messy business of ever investigating even the tiniest thing. It’s astonishing how many people were apparently just waiting for the first chance to give up on doing anything at all for themselves.

Capitalism.

Some people are smart enough to be a lawyer yet still really fucking stupid.

How like your wife does she need to be?

I also doubt ChatGPT is just going to say “vote Republican”.

Propaganda is usually far more subtle than that. You can gently color the way the issues are framed, choose what to present as the important questions and deciding factors, omit to mention certain evidence or arguments, describe some candidates in appealing terms and others less appealingly, etc. All it needs to do is nudge some of the questioners a little to the right, not overtly or dramatically, but just enough to tip the votes statistically just a little bit towards the Republicans.

Well the tech CEOs are doing very well for themselves, governments are finding it easier to suppress dissent than ever, and corporations have a ready excuse for firing more people and paying the rest less. Those are the benefits.

He didn’t say they’re benefits for us. There are plenty of benefits for Sam. And we’re a sacrifice he’s willing to make.

And I, too, am embarrassed on behalf of the tech industry — by our collective failure.

It’s not usually failure though: it’s success in something that’s deliberately hostile, abusive and exploitative towards users. Failure world be to try to serve the user and get it wrong. But modern tech is designed to serve capital, not users, and on those terms it is succeeding.