imabotbeep2937

In short, the latest versions of LLMs now include many tricks which have long been known to cause confirmation bias in humans.

The reason 4o had a sexy female voice, and giggled, isn't to make it more useful as an assistant. It's because a myriad of psychologists and addiction experts have shown those are the easiest ways to fool people when you're looking for a few billion more VC.

Seeing people trying to applaud LLMs for being queer... When they're just really bad at forming coherent stories, worse at sticking to facts, and massively manipulated because unregulated AIs always start to look like 4chan fast...

Sign of the times.

"Against traffic" is different as a bike or led though. Sometimes when lanes are very narrow and visibility isn't good, you need to go against traffic (on the shoulder) so that the driver sees your face and responds faster. And you can see to bail if someone hasn't made room.

But if it's a Chicago intersection and you just said I'm on a bike all 12 lanes have to yield to me and you zoom diagonal in front of cars... the Darwin Awards are never short of applicants.

E-bikes and scooters and stuff have made this 1,000 times worse. The number of drunk kids who think they can go max speed down a hill through intersections and cars will just magically make way for them... Terrifying.

Lane splitting motorcycles and bicyclists are literally always illegal btw.

Lane splitting is for air cooled engines stuck in traffic, where they literally need to move or shut down. (It is technically illegal to park the vehicle with engine shut off in case you can't restart). It has absolutely never been legal for anyone to shoot through gaps in traffic because of erectile dysfunction compensation. It is only in the absolute emergency that you need to walk your bike around to keep it from melting.

In general I do agree. Campaigns to "see motorcycles" just end up with 80yo biker gangs thinking it's okay yo blast their engines in residential. Campaigns to "share the road" lead to bicyclists who will absolutely charge an SUV just to get on the news. I'm all for road safety but it absolutely requires bike clubs of all types to stop being twits.

Literally in the article.

Key missing component: CAFE needs a rewrite. US emissions laws have a cutout so that larger vehicles have less stringent emissions requirements. The problem is that this is no longer a gap it's a chasm. Slight hyperbole but a Japanese kei car many Americans would love town own muat do closer to 50mpg, while a monster pickup can do 15mpg. It ends up that a small car can cost $15,000 and the giant pickup costs... $15,000. Many consumers compare the two and wonder if the econobox is really a good choice.

The bright shining hope: Most Americans do not want these big cars. Legislation is making small vehicles less competitive, when it should be the other way.

AKA - we have good, cheap(ish) electric trucks now. Legislation doesn't need to worry about farmers who can't afford electric anymore.

Posted article quality is not always very good here lately.

Clickholes get too many votes.

"I have run production websites where I didn't patch security for months or years on end." Linux users wondering why nobody takes them seriously.

Most modern distros have this out of box.

Lot of Linux forums need to update their assumptions by about 20 years.

To be fair. All good distros can run from a USB stick or whatever now. The use case for a "liveCD" is limited. Puppy Linux would just be a toy today. It used to matter to me in the days of slow internet, limited storage, etc.

Now just grab Linux Mint or whatever. Use a "real" distro with a community. Install it if you like it.

Geeky Linux forums tend to be people who haven't given new distros a chance in decades, and still think Gentoo and RedHat are the major players. (Desktop Linux it's Ubuntu, followed by Debian and CentOS).

Anyone who is overly zealous about anything is always wrong in the end. Including testing.

"Why would people mock everything? Why not stand up a real test db and test on it?" Because the test zealous have explicitly declared that EACH test should be atomic. Yes you can find these people at major tech conferences. Each test should mock its own db, web service, etc. Every single time. And it should do that in no more than a few milliseconds, so that the entire project compiles in no more than 2mins, even for the largest and most complex corporate projects. And these tests should be fully end-to-end, even for complex microservices across complex networking architecture.

Some of you may be rolling on the floor laughing at how naive and time-consuming such a project would be.

We all agree such testing is a noble goal. But you need a team of absolute geniuses who do nothing but write "clever" code all day to get there in any sizeable project.

My organization won't hire or pay those people, no matter what they say about having 100% coverage. We just do the best we can, cheat, and lower the targets as necessary.

This is true of literally everything in the new economy.

Internet? Wait until the moment your "promo" cost ends and your bill goes from $80 to $150, threaten to quit, oh wow magically you can have $80 again and a free mobile phone line.

Any subscription service is like this. I sometimes grab a Blue Apron when it's 65+% off which is anytime I want. My ex used to do this with clothing subscriptions, up to 80% off.

There are laws against things being "always on sale". But now they're just being used to punish lazy customers who don't keep up on their promos. Only lazy or ignorant people pay the "real" price.

Oh hey would you look at that, another billion dollar IPO with no plan for profitability went bankrupt. Weird.

Presumably they mean gig economy aka artists are vastly undervalued.

For instance. It's not that AI is replacing artists. It's that people think you don't need to pay a license for generated images, even when they were clearly and provably stolen from copyright material. The bar was just lowered. If "AI" is used to remove the watermark from Shutterstock people think that's legal now.

So WHEN gig economy workers get picked up by a company. Yes they pay for a software license as a "tax" on going pro. But from personal experience. A vast amount of art and content is made by people from developing economies on Fiver or whatever. Many of those licenses are stolen.

And now everyone thinks you don't need to pay artists anymore. So nobody will generate licenses.

Adobe was basically right. They're just going at it in the maximally enshittified manner.

Baba Is You would be perfect for teaching people security, penetration testing, and so on.

Lateral thinking puzzles are underrated, and orgs mostly punish lateral thinkers. Even though the lateral thinker is going to catch way more bugs. They will also generate more bug reports by doing so, and that's bad for numbers, stop that.

Or the old FANG hiring problem. "How would you implement a binary search tree for this problem, and do it only on a Google doc, no auto complete."

Good programmer: "I would never do that. That sounds horrible. Who thought of this?"

Opportunity cost.

What about Go, or Checkers, or Call of Duty, or...?

That's the issue. Yes sitting down and concentrating on difficult tasks for long periods of time is good exercise. So is jogging, which isn't sitting down. Or rowing. Or lifting weights. Or...

Chess probably has much less benefits than learning a musical instrument, for example.

DuoLingo is deceptive at best. It's made addictive at the outset so you feel good. You feel like you're learning. But long term the platform has what is now dubbed enshittification. It gets exponentially harder to keep up with reviews, there are known problems with the lesson structure that will never be addressed, etc. And if you pay money, they'll make it easier for you. Hm.

Real critique example: Duo relies on 1:1 translation. Native language learning isn't even close to 1:1. Every language learner hits a point where they can say, "I know exactly what this phrase means in the target language, and there is no direct translation in English". To finish a language in Duo, you will have to memorize dozens/hundreds+ of totally misleading and bad translations like this.

Tldr. Selling kids DuoLingo for language learning is like selling kids cigarettes to help them learn to build a fire.

Demake in modern times is often used for example with porting DOOM to (insert electronic device like a temperature sensor that shouldn't normally play games). Since the target platform has less features, it's not a re-make, it's a de-make. This could also be for more academic purposes, like "what kind of game would we make if we deleted all the guns out of DOOM".

But it comes from the old days of gaming. You could play an amazing fighting game at the arcade, but the home console had considerably less power. So the home version of that game might be very limited.

Back then we would just call that a port. Port was used to mean taking something from its native platform and putting it on something else, which was rarely an upgrade. Putting an NES/famicom game onto Genesis/master system would involve upgrades (sprite capability) and side or downgrades (sound chip differences).

Port today generally means a game has the same quality across all platforms, barring the bare facts of hardware capability. So de-make now often means a port with some intentional limitation. "I know a digital home pregnancy test can't run DOOM, but can it run enough to be playable?"

People naming distros like it's 4chan from 20 years ago. Redhat and Gentoo. Oh my.

Linux Mint, Pop, and more, all now provide "Windows without MS" which is what most users really want/need. Even with gaming now that Steam has multiple ways to run on Linux. You can put grandpa in front of some Linux boxes today and he won't notice the difference.

Debian, BSD, etc. I totally get you want the most stable thing going. You like to pretend that memorizing arcane spells for 20 years is the same as intuitive. I get it. I run some remote stuff on a BSD box. I don't know a single serious command without a lookup. But it's stable. I thank the guy who made a cheat sheet for common commands.

Even on the security front though, the tinfoils are doing weird containerized OSes and stuff now. Opening a browser tab can spawn a whole sandboxed OS in seconds. Your pitifully out of date Debian repo doesn't really stack up anymore.

Basically. You can run a modern OS that doesn't suck to use. And that OS can be Linux. With Windows 10+ poised to drive many away from MS, distros need to get with the times and be ready for the masses.

Or. "Serious users will move to X" is a bad argument from the 1990s. Newbies have an OS that pros can enjoy. And if I want to run Debian, I can fire it up in a sandbox or whatever. It's the 21st century. It's not like I have to wait for physical media to spin up to install a live copy.

It's both. Especially the out of context tinfoil rage response. They always existed. But now it's so common to see some totally benign article about pizza, and the top comment is "don't let them tell you not to remember in 1995 when US implanted radios in Syrian babies".

The algorithm is being trained solely for engagement. It is horrifying.

Agree. Content was the OG slop. Buzzfeed with monkeys on typewriters.

The problem is that dopamine addicts generate outsized engagement. I know a literal crack mom who spends a solid 90+ hours a week watching accident videos to keep her brain triggered. The algorithm caters to her. Send promotional emails daily or more, constant notifications, recommend the same few videos over and over. Gotta get in there before she clicks another car crash video.

IMHO: Marketing is a top societal evil right now. If the media machine wasn't so desperate for content, AI wouldn't be a fraction of the problem it is. But with everyone obsessing over the next piece of content, fake AI presentations are mandatory.

The phrase "Eliminate corporations from academic research" does a lot of heavy lifting here.

If a company wants to bring a pesticide to market. We would like them to bear the brunt of safety research costs before they launch. You might say that product safety is different from pure academic research.

We're just arguing semantics and incentives at that point. Like giving them a tax break for that research. Sure that sucks. But how else shall we make it happen.

The whole world will be headless content. There won't be any web pages, or bank sites, or TV networks. Nobody will be a developer. We'll all just be content authors, like Google Maps Guides basically being unpaid interns checking restaurant data for Google.

You open your phone, it just shovels content. And it does absolutely nothing but optimize on addiction.

No apps, only masters.

Nobody is realizing this coming singularity.

Your phone won't do anything else. For 99% of people, they pick up their phone, AI will just decide what they want to see. And most will accept it.

Someday everyone in the room will all pick up their phones when they all ring at once. It will be some emotional trigger like a live feed from a school shooting. Everyone in the room will start screaming at the totally different experiences they're being presented. Evil liberals, clueless law enforcement, product placement being shown over the shooter's gun. You'll sit horrified because you returned to a dumbphone to escape.

That will be the reality if this AI assistant stuff isn't checked hard now. AI is getting better at addiction an order of magnitude faster than it's getting better at actual tasks.

"brother didn’t bother to check the flight code I sent him via message when he asks me when I’m landing for pickup"

Yeah but what about people going to the wrong airport, or getting scammed by taking fake information uncritically? "Well it worked for me and anyway AI will get better.". Amen.

"I've seen people get rich off crypto" and I've seen hundreds of gambling addicts lose their house and put their family on the street when the bet didn't go right. Even when you're talking about serious coins, ignoring the rug pulls. That money comes from somewhere. Who's holding the bag?

Crypto people tend to have an infinite dose of "f** you got mine" even if they didn't get theirs.

I've been tested to 22+kHz, autistic, and older. It was the worst era.

Some of the awful roars of noise I've heard while older people called me a weirdo. CRTs, early checkout scanners, some audio systems.

And back when "as seen on TV" electronic dog whistle trainers were all the rage? Torture.

Feynman was not a great teacher. Lots of people love his lectures now. But they're not getting graded on it.

Actual students were considerably more mixed. Obviously individuals vary so if I post how most of his students were totally lost in class, they'll post how one student went on to win a Nobel prize and say gotcha.

"Easier to learn" is abjectly false.

"Easier to get code submitted to a code base" is only marginally correlated with learning anything.

You're way, way under thinking the problem.

People who work at your bank right now are pasting your personal details into language models and asking if you deserve a loan. People will figure out how to get this data back out.

"But the model only uses old training data" there are myriad lawsuits in flight where these companies took information they shouldn't have, in all forms. Prompt engineers have already got the engines to spew things that are legally damning.

A real hack, which might archive user inputs as well as exfiltrate training data. We're only beginning to imagine the nightmare.

Can people use these language models to get private Google info? Only if Google was dumb enough to include that in the training. Hint: Yes, it's much, much worse than anyone imagines.

To take a stance much more suited to this forum. It's a massive travesty that without some work Americans are now redirected to BBC.com, which curates news to be American-facing, and thus IMHO bows to advertiser pressure. We can't see BBC.co.UK, so we can't know what the other side of the news even looks like.

All news services do this. And it already fragments and destroys and hope of really talking about the news in a healthy manner IMHO.