ticulatedspline

I get your sentiment, though the problem is older than AI, the convenience and market penetration has just ramped up the visibility in this arena.

Many books pre-ai were garbage but not at the rate they could be today. If anything people operate on greater discretion now than in the past because even the worst garbage still had a non-trivial barrier to entry, either sitting down to regurgitate your own garbage or paying someone to make some garbage for you.

There's also some historic precedence of this, the rise of Photoshop/CG and digital manipulation certainly led to a "is this photo real" to similar effect despite manipulations going back to the origins of photography. That acute concern was not as prevalent in my mind before editing became common and accessible.

Is browser compatibility that durable? I would think that both java-script and CSS would be subject to some level of compatibility rot against a browser from 2056

I prefer leaning on Hanlon's razor for things like this, though even in ignorance the effect can be real and dangerous. That said I would not put any model from any country past an "Order 66" kind of thing

https://starwars.fandom.com/wiki/Order_66

> Its a decent job and some one has to do it

someone has to do it but it's not a decent job. Content tagging and content moderation are basically the "ditch digging" of the white collar world and often comes with significant toll on those exposed to such content continuously.

nah, you'll never get there that way. Even if there was willpower to do it legislating these things is too hard.

What we need is some greedy lawyers and civil liability. Find a campaign that infected enough individuals, fire up a class-action lawsuit against the website on which the ads where actually shown (not the ad provider).

If they win the landscape will change. It would set a precedent for civil liability for infecting your users even when it's not your content directly. That would likely spook the shit out of the market, security becomes a feature.

good study with, entirely expected results? Variances in bioavailability are well known no? A block of wood has calories but you can't live off it. As for processing there's been papers describing "roasting meat and veggies" as a gateway to our larger brains because cooking makes calories in food more available.

Does a cooked steak and a raw steak actually show different values when measured with a calorimeter?

their chosen dishes also seem to conflate a lot of factors, like processing, vs flavor. it's known that artificial sweeteners can elicit insulin responses despite not being a digestible sugar, I've heard studies showing that "Tasting" in the gut can perceive the sweetness.

> The disclosure conversation—”I used AI for this”—is starting to sound like a cook announcing that the kitchen has an oven.

Think OP misses the point of that disclosure, It's more like the cook saying "I ordered out", Ie, they had nothing or very little to do with the output. Which is a good disclosure, now if an employee does own the output, then there's no or little need for the disclosure because for all intents and purposes it's content they've approved as their own.

This is also not a new phenomena, if you had a senior (whatever) with 15 years of experience and you asked for some thing, and they in turn delegated it to their intern they would either review and own the output from the intern or if it was something that could be passed on with only cursory review likely qualify that it came from said intern, which both gives credit, gives blame, but also tempers the consumption such that the recipient knows it's not directly a product of the senior (whatever) with 15 years of experience.

the only difference is now even the interns have interns everyone can delegate any task.

> Any project that sees a semblance of success will be cloned overnight many times over and will find itself competing with its own copies.

wasn't this always a risk with open source? also isn't that part of the point of open source? undoubtedly there are some forks out there that have become the new baseline. Besides, survival of the fittest. If some vibe-coded clone can compete, can manage updates, win market share, produce value superseding the original does that version not deserve it's share?

Personally I don't think it's all that trivial to maintain a large FOSS project, that vibe coded clones will simply not keep up because they were low effort and when they realize even with all the AI tools in the world it's still work to maintain a product they'll die off.

> what motivation remains to continue publishing as open source?

What motivation was there ever? prestige? money? accolades? Is that motivation actually gone?

Also closed source is not a protection anymore. look at Jev, they didn't even release the weights but everyone and their mother has a Jev-like model either out in the wild or in the works.

Aside from being a drop in a sea of games, first impression from the store page is pretty lackluster.

The character is, like a glass mug? not a lot of personality there.

The first video doesn't show a lot of interest, there's like 6 cuts in a row with the cup moving right and up making the gameplay look repetitive.

Overall look/feel is still very "Unity demo", a lot of what makes the greatest games great is not just their gameplay but the feel and details. Imagine portal without GLaDOS and a low poly un-textured style but exactly the same puzzles. It never would have made the mark it did.

Like I'm looking at the second video and there's these avocado things, are they bullets? enemies? looks like the cup has some large radius, fast orbiting, ball that just erases the avocados. like they don't splat or react, your cup character just lifelessly floats around and stuff is happening, it makes the fight feel sterile rather than engaging.

Funny, I was just thinking the other day that a Watergate like occurrence today would be such a non-event for the current administration.

Not sure how old the author is in the story, I imagine them at like 8-11 ish. I don't have kids so maybe my view is too narrow but of the peers I do have with kids near that range I don't think I know any of them that have a "computer".

many of them play minecraft but as far as I'm aware it's entirely on tablets.

Granted the vanilla experience offers lots of exploration and creativity beyond nearly anything I grew up with but it's still not the same feeling of roguish discovery described by the article.

Heh love that it has the star-field screensaver. nice touch.

I lament that these kinds of rogue explorations are less and less available to kids these days. Sure in many senses there's much more in the space to engage with, powerful tools like scratch and now LLMs to answer every question but I suspect most kids today will be nostalgic for some mobile game they have no agency to explore outside the rails or some online game that will ban you for cheating.

Contrast that with experiences like the author and the taboo joys of rooting around in the files and "hacking" something to do the thing you want.

Unless there's other data it's not clear if or how AI is actually involved here. the core issue seems like something any provider could have started doing long before AI.

Well, it would be weird if a loaf of bread cost less then the wheat it's made from.

> Michel showed me a vibe-coded app he did for the restaurant, which researched every customer beforehand on the web to see if they can offer a personal experience of some kind - a small detail like a joke on a napkin upon serving them.

Goddammit that's creepy as hell. I'm going to have to start using burner phones and wearing a full face mask in public now because every privacy invading, ill informed, short sighted, "bright idea" featured on "Clients From Hell" (NotAlwaysRight.com) is going to be vibed into existence, isn't it?

I think we're going to start seeing a huge avalanche of remakes and mods of classic games. I've been working on my own modding of a classic game I still play and the experience has been both fascinating and surprisingly fruitful.

Like most things the majority of these attempts will be garbage but some will certainly pan out into great new or extended experiences.

Not easily, assuming you even use any apps that are so trivial you could accept a straight up vibe coded replacement, side loading is becoming more difficult

I think the dream will never die, but it might go underground for another turn.

Problem was that broadly there's two categories to compete in. "Superior product" , and "cheaper". the last round of artificial meat was neither.

I genuinely believe that eventually we'll have a proper "biological revolution". much like many inorganic technical revolutions have shaped the world (AI being one of the most recent) I think we'll have a breakthrough that opens the doors to the kinds of biological technologies from science fiction.

At that stage I would anticipate an assault on both categories. Ground beef/chicken/pork of sufficient quality that undercuts factory farmed meat. But also superior products. the best wagyu in the world is just tissue, a collection of cells. I can certainly imagine a future where chunks of meat at that quality or even better roll out of vats. softer and more marbled than any real meat because it's not limited by the conditions of being grown within a functional animal.

And when it's better or cheaper people will buy it, and squeamishness will abate as it becomes normalized. that the meat may be more humane will be a nice value add but never enough to prop up the product itself.

> "éš-dam", can also be translated as "brothel"

My own personal guess "I cannot see anything" may mean "blind". door could be a choice of brothel rooms. so basically

"A blind 'dog' enters a brothel and asks, which hooker should I take?"

For those only consuming the title. yes, no duh humans have access to your conversations and read them.

The lawsuit is targeting a gap in the privacy policy that doesn't explicitly call out contracting out data-labeling and model-training tasks to 3rd parties.

As far as I can tell there would be no lawsuit here if they were using OpenAI employees to do the evaluations instead of contracted labor. seems the entire suit is hinging on this single gap:

> the Privacy Policy lists eleven kinds of outside companies that receive users’ personal data. They include hosting, payments, customer service, analytics and identity verification. None is a data-labeling, annotation or human-evaluation vendor.

"Promise" is a strong word here.

> According to the complaint, the Privacy Policy lists eleven kinds of outside companies that receive users’ personal data. They include hosting, payments, customer service, analytics and identity verification. None is a data-labeling, annotation or human-evaluation vendor.

Would a "Regular" person interpret the policy such that the described party was one OpenAi "promised" not to share data with?

AI poisoning is currently the new exploit frontier and I don't think it's going away anytime soon.

I was just contemplating the other day that agents we use at work are exposing new attack surfaces. For example, historically a folder of documents (maybe google docs) isn't a huge risk, particularly if it doesn't contain sensitive documents.

Now though if one employee is running a harness capable of computer control, and running an agent that reads from that directory, simply dropping some files with instructions could poison the AI to leak info or even own the host.

Skills are another supply risk, malicious instructions could be added to them.

finally paying up for chatGPT and using the Codex desktop app was my real "I'm sold" moment with AI.

Setting up projects and working with the AI on local files has been great, but only for my personal account. I've been trying to get it set up for work that provides OpenAI models through a 3rd party tool, company hosted models, as local-machine models and the UX is just straight up awful.

there's no GUI for profiles or custom endpoints, the config.toml sucks and the overall experience is primitive.

At this stage I really want a Codex-like harness but I need more fluid control over the models, I want features like pinning a project to a provider, as well as pulling in all the models from that provider, also having all providers available.

So if I need to pop over to one project to consult about product A, then pop to another project to do some code analysis on product B I can do so fluidly and have my tokens billed to the right place for each concern.

or a project that can span all of the resources. like having the OpenAi models orchestrate sub-agents on the local or hosted models.

It doesn't take a lot of bad people to subjugate a lot of good people.

The read was very interesting, and I think you're right, mathematicians are still very much needed, in some sense now more than ever.

But the effect is the same across other professions. We need mathematicians, we need professional grounding but we only "need" experts.

Every step AI takes feels like it slices people off the bottom of a profession, making them less relevant, while simultaneously increasing demand for the top.

the vicious cycle I see is it will "Eat the middle", you need people with juuust enough acumen in a subject to drive the AI, to set a goal and nudge to a direction, and you need a real expert to look at the results with a critical eye but you don't need the middle of the curve. I think over time this will depress hiring and wages of new people, and stunt their progress towards expertise by destroying the middle ground. Then when our current experts retire we'll be left in a lurch.

> Computing has historically been a field of wizardry.

Counter argument, "historically" is not last year but closer to 2 decades ago.

modern computing has sat atop a mountain of sorcery for ages. You may be able to squint and call a good compiler a "tool of power" more than sorcery but there are plenty of other tech stacks that simply sit upon one opaque box after another.

And it's not unusual for sorcery to produce crap, even if it's production worth crap. Dreamweaver had a WYSIWYG editor ages ago and as someone who was hand crafting HTML in that era it's code was typically a mess.

Now that's the norm, almost nobody makes HTML by hand, some js framework does. And it's often shit, I've seen entire sources of nothing but <div> nested ad-nauseam, conjured up from the depths of "I-don't-care-as-long-as-it-works"

As an old curmudgeon I recoil at any service I can't interact with sans-phone. I hardly use mine for anything.

Just be glad your phone number is tied to your account, if you lose the number itself you are royally screwed, and basically anyone who gets your old number owns all your accounts now.

Isn't there an SMS option? buy a new phone and have them send you an OTP via SMS.

> once you're phone is gone, you are completely over with society?

yes, I'm waiting for the Black-Mirror episode where someone drops their phone and basically dies, can't contact anyone, can't unlock or start their car, can't get an uber, can't pay for anything, can't navigate without gps, can't log into anything etc.

Can you provide a compelling argument for why it should be illegal?