So not half-staff?
So not half-staff?
Staff.
This is the future of all media that Silicon Valley is pushing for.
Liberation day.
I think almost all of my bag checks involve either sporting equipment or beverages of some sort.
The real solution is to just pack light.
For that matter wary of what?
Lmao www.peoplespolicyproject.org/2018/06/11/d...
In other words, they are storing customer personally identifiable health data. This is *not required for reasons other than monetizing that personally identifiable information*. A reputable product would be e2e encrypted and anonymized at its very core.
Someone tell JFO Patrick he's not white. bsky.app/profile/mari...
This is the perfect bluesky outsider post.
Not exactly going out on a limb here are you?
Also we get to let Elon grift by pretending we're gonna fly around in tubes.
They're all lunatic bots. It's ok.
Gotta save up for ICE signing bonuses.
370M Americans.
They have every incentive to hype the most glorious case and they come up with this?
In a context where I owned my data (and use of it and its derivatives was exclusive to me) I could see some cool uses.
Knowing what we know now...other than a 100% on-device or self-hosted option, or *maybe* some sort of e2e encrypted solution that you trust the tech to jold up...why would you willingly touch any of these devices?
Just for fun imagining this in the maybe 3 places where it woukd actually be at all questionable.
Or to be more explicit - the GOP Southern Strategy has been a/the driving force in electoral politics for as long as most people on this site have been alive.
There are a lot of layers to that. It's easy enough to just track campaign materials, 70 years of televised debates and their coverage, media coverage, etc. But the balance between voting decisions and how campaigns operated is going to be a bit more subtle.
The red flag for me isn't that LLMs can't do basic math or count letters. The red flag is that companies that have sunk billions upon billions of dollars into them haven't given any indication that they've been able to figure out how to divert these inquiries to tools that can handle them reliably.
Did Sam Altman write this?
It takes an incredibly aggressive level of self-delusion to pretend that ex-twitter is anything but an opaque black box built solely to advance Musk's agenda and wash it with the faux endorsement of whoever participates there.
I think the author forgot the word "claim" here. Reporting *in general* seems to leave that off waaaaay to frequently when dealing with bad-faith actors.
You don’t say.
Whoever expected that from this was delusional.
He's describing ding dong ditch.
That might be the case! It's fucking word salad that avoids communicating whatever the actual bottom line is.
Max Beans.
(Repeat for 5000 other issues, large and small)
Reporting on stuff like this needs to be so much better. As it is it's just a bunch of words and numbers that may not be consistent with each other. It might be a sensitive but noisy test, or it might be complete shit. The article is an abject failure at clarifying that.
1/3 (true positives) * ~2 (reported positive rate) = ~2/3 - i.e. fewer detections on top of all the noise, assuming they actually used consistent terminology in the article.
There's always money in fraud.
"Over double" the detected heart failures vs a doc with ears and 2/3 of them false positives does not math well. No idea if they do any better elsewhere but I assume everyone quit using them for a reason.
Much moustache. Very revolution.
Don't worry it's only a temporary limitation.
This is extremely funny to me
It's absurd we need to waste reporting effort confirming what we all knew.
@patcasey.bsky.social you have failed us.
This is, I swear to Christ, the lead story at the Atlantic Monthly right now
But Ben if that lane didn't exist where would the closeted nazis hang out?
We're just out there laying the groundwork to destroy art and culture. This crap normalizing prompt engineering as art came on-screen while visiting family and I just about lost it. It's just as bad as it sounds. www.hollywoodreporter.com/tv/tv-news/p...
I suspect that would be sentiment among people plugged in to MRR...but by definition that's already outside the mainstream.
Realistically his description feels like it would have applied most anywhere in the US mainstream with the exception of people even recognizing the Crass logo.
It's a tough problem. On one hand you absolutely need to set and enforce trust based on knowledge (that normies lack!) and standards. On the other hand delegating that control to monopolistic hegemonic surveillance capitalists does real damage. Ultimately most phone apps are borderline malicious.
Beyond orgydome.
Stanning for MN-grown WMDs.
Also you can maintain this tradition with immich.
Every day you can look back at memories of who threatened to beat you up 5 years ago!
What do you mean by "more with Courtney"??? bsky.app/profile/cour...
The best toast is woke toast.
It was stupid (and obviously so) then and is stupid (again obviously so) now. Even if these things actually end up being important...how people interact with them is in flux! The only novel part is surrendering your agency and letting the machine think for you which has zero place in education.
We really need to ditch the tradition of media credulously and casually parroting big tech propaganda at every turn.
Daily reminder. When an LLM replaces thinking and infirnation processing you gain nothing and lose everything.
If you're the sort of dipshit that watches a TED talk and thinks you've gained PhD level knowledge, or think ChatGPT is a PhD in you pocket - maybe 100% video is for you. And some topics and learners 💯 suit video. But in general video is about CONTROL and MONETIZATION, not about education.
Videos and podcasts are unambiguously 1000x worse for vast amounts of learning - espesially anything with real depth - and as reference than e.g. text with selected a/v content. But we don't get that online anymore because it's easier to monetize content you need to stream at soneone else's pace.
Great to see anti-literacy education activists!
They're honestly great other than that they kill you every once in a while.
Honey...we're gonna raise the Worldwide Leader!
It's not even an obscure thing! The GOP impeached the sec of homeland security like 18 months ago.
Slavemaxxing
What we really need are good foods with good bacteria!
Understanding how people misuse and mismarket a technology doesn't change that LLMs don't actually work in this application - especially not sycophant chatbots like Claude, ChatGPT, etc. Sure people do this, but when they do so fact and fiction mixes freely and they forget what they even wrote.
Directly to the point: Misusing LLMs as "search engines" damages cognitive ability and information retention. In this study people who used LLMs to do the work *literally could not remember what they wrote*, likely due to never processing the information in the first place bsky.app/profile/matt...
LLMs are not search engines. This is *critically important* to note. They're statistical random token combiners that cram together plausible combinations of meaninglessness in a manner that's engineered to look plausible (but may be 100% false) and make the human happy + lure them to do more.
What kind of brainstorming partner? A person. That you develop comradery and rapport with. You grow your knowledge and experience together.
Monkey's paw gonna curl real hard on that.
Also root veggies can be superb.
Like a SV steak is still a steak just ostentaciously uniform in the middle. But chicken breast cooked to like 140F without salmonella risk is really great. As is chuck roast cooked at say 133-135ish for 24 hours. Unlike a SV steak these are fundamentally different from traditional preparations.
The thing is...there's almost no learning involved if you actually understand cooking, save for an expanded range of "safe" and a wider sweet-spot. If anything there could be a deskilling risk tbh. It makes a bigger difference on roasts, chicken, etc. because you can cook completely differently.
The whole thing is just an industry-funded marketing whitewash of a net-damaging product.
We are all juice packs.
I'd definitely love to stop seeing them...
2 completely different frequently advocated by the same people.
Craig missed the opening of the fair, but she did make it there to pass around the sweet marthas. racketmn.com/angie-craig-...
Not just dystopian novels either!
I'd also suggest that if the code is expressive that connects intent to its output, pushing cases like this beyond mere negligence.
I'm half convinced that this is the current policy approval process.
Sounds like a call for a general strike in tech at this point.
Soon you will yearn for the good old days. fortune.com/2025/01/17/w...
He's also *thrilled* about Trump's health.
I finally read computer scientist Joseph Weizenbaum’s 1976 classic “Computer Power and Human Reason.” This book deserves a massive revival in our current age of grotesque and largely thoughtless AI creep into everything:
It was the headline result. After working with the AI for 3 months, doctors missed over 20% of the stuff they had caught previously. This isn't surprising or necessarily catastrophic, but it is something that needs to inform how we'd train doctors and use the tech.
Yeah obviously none of this is going to "work" in any way that involves better outcomes for anyone.
You end up needing to structure both training and application in such a way to minimize deskilling effects in order to get high-quality and sustained results.
It's more subtle than that - the AI combo catches more because the AI is inhuman and catches abnormalities a person wouldn't, but also misses stuff a person wouldn't. In the study even expert humans got lazy and their skills atrophied after working with the AI.
These are "good" uses of "AI" (which is nothing like the stupid chatbots) with the caveat that the bot alone isworse and the doc's skills over time can easily erode if not relied on until you are worse off. Because unlike a calculator it's a black box. It's the "good if handled carefully" corner
This is their hammer to push through age (actually ID...) verification laws.
Disrupting reality by automating Jonestown at scale.
LMAO who could have seen this coming? bsky.app/profile/matt...
I dare you to go to Annunciation School and tell our grieving community, in effect, guns don’t kill kids, antidepressants do. Just shut up. Stop peddling bullshit. You should be fired.
To be fair if you read between the lines on that article you might infer that O'Malley was shadow SSA commissioner or similar and he has been making national appearances (e.g. Arizona and Iowa this month) pushing that topic. But it takes much more than that to meet the moment.