otter

It seems like a chunk of people (and large institutions) are trying to 'wait it out' and ignoring all the awful replies / content

Once people switch, they're unlikely to go back. Problem is getting them to put in the time to switch.

I'd call it a win when we start seeing "____ said on Mastodon" in articles

any tips on which models you recommend?

Sounds a bit more like the format is getting very exhausting to people, and more people are recognizing the unhealthy effects.

So now they need to go the opposite direction to try and fix it

I'd say if the action starts to affect those trying to use the internet for other purposes.

So if visits to X affect people's ability to do Y, then that might count. This would depend on scale (ex. a town might get overloaded more easily) and reasoning (ex. both X and Y used the same bottleneck service)

Maybe?

It can be abused, but weighing the pros and cons it can be worth it depending on the community.

We had it on a subreddit and it worked well the vast majority of the time. I only remember one case with abuse, and when that happened the abusing user got banned shortly afterwards.

Maybe the larger American owned projects are located in places where energy is cheaper / cooling is easier

It's still officially called a recall because it's something the government asked them to fix, but yeah, it's super misleading to call it that

Ah gotcha ok, thank you!

Do they need to be driven anywhere for the fix or is it like an automatic update?

Someone mentioned in another thread that this is still technically a recall, but I didn't understand why.

I felt like calling it a recall might make people distrust future news about issues (or a bigger more traditional "send the car back for physical part swap" recall). Maybe a different term is needed?

Yea I share the same concerns about the "AI", but this sounds like a good thing. It's going through footage that wasn't going to be looked at (because there wasn't a complaint / investigation), and it's flagging things that should be reviewed. It's a positive step

What we should look into for this program is

• how the flags are being set, and what kind of interaction will warrant a flag
• what changes are made to training as a result of this data
• how the privacy is being handled, and where the data is going (ex. Don't use this footage to train some model, especially because not every interaction is out in the public)

Body camera video equivalent to 25 million copies of “Barbie” is collected but rarely reviewed.

I get this is done to hook the reader, but I feel annoyed how these comparisons don't actually make it easier to visualize the scale

For around $50,000 a year, Truleo’s software allows supervisors to select from a set of specific behaviors to flag, such as when officers interrupt civilians, use profanity, use force or mute their cameras. The flags are based on data Truleo has collected on which officer behaviors result in violent escalation. Among the conclusions from Truleo’s research: Officers need to explain what they are doing.

I guess that sounds like a good thing? Flagging the videos instead of waiting for complaints

:)

Regular breaks would be best, especially if you can go for a little walk around. Another thing would be contrast, so it's better to have the screen at a similar brightness to your surroundings. I'd also say that distance from the screen makes a difference, so being further away (with bigger text) would help.