DreamerOfImprobableDreams

Good points, but even so I’d still rather have a less-than-optimal sub that still lined up with what the characters were saying in the dub, than having the two conflict with each other. (Maybe if you turn the dub off, the subs could be replaced with the better sub-only version?)

Yeah, to be honest I’m a bit worried studios might use this technology to start changing lines in already released movies, so subtly only film buffs who know the film line-for-line would ever notice.

Even done for well-intentioned reasons (like removing slurs from old-timey movies) it would open up a whole moral and ethical can of worms about whitewashing the past. And I have zero faith in studios to do it for well-intentioned reasons-- especially not when they’re under pressure by religious fanatics, authoritarian governments, etc.

Also, for some reason the dub and subtitles use different translations, so the character will be saying one thing and the subs will read something completely different. It’s confusing AF. Often end up getting so frustrated I just turn off the dub and focus exclusively on the subs.

I remember hearing about a high-profile case where the AI would dock points if someone’s resume listed them as participating in women’s sports as an extracurricular, while giving extra points if it listed them as participating in men’s sports.

Also, bias doesn’t necessarily have to come from happenstance. Unfortunately, humans tend to have unconcious (or, sometimes, not-so-unconcious) biases against women and people of color. There was a study where researchers sent identical resumes to a random group of recruiters-- but half of the resumes had a male name and half had a female name.

They found that both male and female recruiters were more likely to rate the resumes with the male name higher and be more likely to recommend they be advanced to the next round of interviews. IIRC, similar studies have found similar results if you give the resumes a “Black sounding” name versus a “white sounding” name.

So if you train an AI on your own company’s hiring data-- which is likely to be tainted by the unconcious bias of your own recruiters and hiring managers-- then the AI might pick up on that and replicate it in its results.

Yeah, and the author all but freely admits that. But they’re apparently willing to throw all that away to have it “feel like everyone’s there.”

Maybe this is me being uncharitable, but to be blunt it sounds like he just wants to feel like he’s sitting at the cool kids’ table.

Did people’s parents not teach them about putting things on the internet they didn’t want shared?

They used to, then social media became a thing and they stopped. Suddenly, it was normal to put your entire life up online for other people to see, and if you didn’t feel comfortable doing that you were the weird one.

My rule is, never post anything you wouldn’t mind the media tracing back to you IRL and then making the top story of the day in your country. Because, while rare, that does occasionally happen!

Also, Chat GPT at least still writes at the level of a somewhat talented ninth grader. Its prose is stilted, and the way it structures essays and stories is super formulaic.

It’s absolutely not at the level it can replace a talented human writer yet. (I have no doubt that day is coming, probably sooner than we think, but it’s not here yet.)

So publishers making the switch will see the quality of their content drop, and with it the number of clicks / revenue they get. Enough to offset the salaries of all the writers they fired? Probably depends on the publication. For clickbait farms, probably not, but the higher quality the readers are used to the more the publishers stand to lose.