The AI angle is new, but from an artistic integrity perspective, studios have been rewwriting stuff for ages. I've seen re-releases under a new director come in with added scenes/characters (Night of the Living Dead springs to mind) even before computers, and Lucas of course inserted all sorts of stuff into what fans would consider a finished trilogy. The effect on me is not one of shock, but rather, "ah, I wondered when they'd realize they can do that...."

Now, all that being said, the big difference here is that they need the actor's likenesses for these new scenes. It's being published with new material, a new plot, and under a new name: that sounds like a new movie. If they need the actors likenesses for a new movie, presumably, even under Indian law those actors are entitled to a new contract, which they can and should refuse to sign. Even if the AI implications of likeness rights are half-baked at this point, it's a big enough risk that no stduio should be willing to attempt this.

Man, if that ain't the thesis of the last 30 years....

If you install a newer Firefox, how do you disable this behavior? Because these ideas are terrible.

Keep your original search visible
Use a search bar.

You can access different search modes in the address bar using simple, descriptive keywords like @bookmarks, @tabs, @history, and @actions
So you still have keyword searches. That's what you've re-invented. We've had that since, what? v0.9? Except now there's an @

Whereas before your address bar was filled with long, confusing URLs, now...
How many times do we have to keep re-inventing these dangerous UX ideas that make it easier and easier to fool the unaware?

Type a command, and Firefox takes care of it
So if I want to know how to delete one or a few downloads, and search for something like "delete downloads", it will now delete everything I ever downloaded? Who wanted this feature?

Cleaner URLs with smarter security cues
So I will know that bankofameric.com is secure in that it uses https, but not in that it's not the same as bankofamerica.com? (That's right, Cyrillic "".)

Let me guess, you also let javascript/WebAPIs change the search term in realtime, because someone somewhere said "maybe someone can use this...?"?

Making single form that 1) identifies wherever the hell you're supposed to be, 2) gives a shallow-to-the-point-of-being-false impression of security, and 3) performs actions that affects your workspace and files is just insane.

Mozilla has moved well past setting money on fire for things no one wants. Now they do it for things I would actually pay money to never experience. ...

Wait, is that their new marketing plan?

the yellow dots ID a printer, but you have to go to the manufacturer to correlate your dots with some serial number. Good for, say, the FBI, but pretty useless a company that owns it. I guess such a company could do a census of every printer they own....

the phrase "watermarking" has been used for this exact application for decades.

Watermarking is damn near ancient technology. They're not even using well-hidden watermarks. At a minimum they could require you to do an FFT or something to produce something readable.

It's amazing to me that any journalist would directly publish any leaked data. That's incredibly stupid. Even leaking the plain text is risky, as you can "watermark" any document by subtly re-ordering words and sentences, inserting typos, etc. which would tell you at least what office the document was leaked from. I'm tempted to blame the collapse of journalism as a profession, because they should have been told this at some point in their career.

Wherever I went with my dad, it was to get A/V equipment, RCA cables, connector for 1/8" audio jacks, that sort of thing. I think Best Buy, Circuit City et al took that business, which didn't leave much to actually keep the lights on. After that I feel like it was all RC cars, and later cell phones (which ny then you could find anywhere).

The movie is the same movie as ever. The only difference is now Meta encourages you to text a chatbot on your phone while you watch. It is truly useless.

The dollar amount matters.

If you're given a choice of $2 vs a 1-in-100 chance of $100 (and attendant 99-in-100 chance of $0), you take lottery. Sure, it's half the expected value, but $2 is so close to $0 that it has negligible impact. Assuming you did the work before being given a choice, you may as well go for the chance at real money rather than the guaranteed pitance. Also, since there's no real risk, there's no risk avoidance.

Also, the more financially-secure (read: older) the less the pitance means. The useful take-away, as they note, is about financial rewards (or lack thereof) biasing survey results.

I would bet it's something more banal. The billable SMS load of this has got to be minimal. It's probably that the backend servers are coming due for a refresh, and some manager finally asked, "why is this even in my budget? Does anyone actually need this? Do we really need to keep paying for it?"

Me too. It was an easy way to get text alerts when an automated test finished in tools that already supported email.

While the goal is always to make money, there's more to consider than the film itself. It's just as often about extending IP protection.

Snow White as a story is public domain. Elements of Disney's Snow White, with all the artistic flourishes such as the dwarf names or her blue and gold dress, is protected by copyright, which is only 7 years from the now very generous 95 year expiration. Snow White as a trademark that you can use to protect lunchboxes and toy make-up kits and such requires a certain degree on constant use to maintain.

If you're a giant IP holder who is also a production company, you probably want to schedule production around some must-have, IP-extending works.

You also see this sort of things in the comic book industry. I remember just after the turn of the century Marvel released four single-issue specials with all their wild west properties you hadn't seen since the 60s. The next year it was four single-issue specials with their horror properties you hadn't seen since the 60s. The next year it was their romance books. And so on. They would like to make a little money on them, and printed to meet level of expectation, but the goal was to have titles like "The Rawhide Kid" appear for sale in a market somewhere at least once in the last half century. After all, Marvel was the company that got the name Captain Marvel from DC just by paying attention to the fact they hadn't done anything with the Fawcett name in decades.

A better solution would be to have the LLM insert comments with bespoke comments or no-op code like "#this code was created by an LLM" or "if {0} { bool __ai_code__ 1; char * __code_source__ "LLM"}"

This is part of what Edward called "the revenge effects of technology". By reducing the time, difficulty, or cost of a task, technology generally does not free up time for other tasks. Rather, those tasks are done more frequently or in larger batches, or the create new, related tasks, and ultimately consume the same or more time.

(One example he gave was spreadsheets. They were done before computers, but were very tedious and thus rare, say once a month or even quarter. Now that they are faster, bosses tend to demand more spreadsheets, and workers tend to spend more of their time on them.)

Private equity scoops up _everything_. Whataburger has been quite successful, the familyjust wanted to cash out and retire.