jarjoura

I would attribute any in-person cultural shift to the generation of 20-somethings trapped online during their teenage years from COVID. They are finally old enough, and have enough power to run things.

The value of an LLM is the dynamic reasoning you get out of it and the cost to execute on that.

I see two forces working against this that proprietary models will always have over an open source model.

1. The biggest is content licensing. Content is quickly becoming gated by systems at the front of their load balancers, completely changing the social contract of the Internet. What used to be a quick google search for recent facts that lead me to places like reddit or twitter, is now completely walled off if you're not physically at your browser and using an IP address from a last-mile provider.

LLMs have pre-trained on the bulk of the information up to 2024/2025, but over time that will be more and more out of date.

Anthropic, OpenAI and Google will all have to pay for access to a lot of this content refresh going forward, and it does make a material difference in the output you get.

2. Liability is the other. A corporation can look at a contract for model access and see one that provides uptime guarentees, content infringement promises and model safety, and pick the contract that shields the corporation from the most liability. A 3rd party hosting platform like fireworks.ai that hosts open weights models won't provide any of that at all. They will simply bill you for time spent on their hardware and make promises that they won't log or inspect corporate traffic.

I'm pretty sure that this is both

1. an artifact of agent tools only doing generic greps based on keyword searches, limiting the scope to a few lines 2. RLHF from Anthropic and OpenAI that bake in catastrophic edit avoidance

I used to have a CLAUDE.md rule that forbid inline comments and only allowed public API doc comments. That used to work perfectly back in 2025 era models. Now these models refuse to honor it and write worse comment slop.

This has become the "voice" of Opus 4.8 and I just roll my eyes when someone puts up a PR that includes more comments than code changes telling a story about all the random things it found during exploration.

These comments do cost us real dollars to endure, both from the model generation cost and the human cost of having to read them. I was hopeful Fable would have been better, but aye, it's worse and writes the same comment slop in multiple places now.

TBF, I just recently picked up this same model, and it's reminding me of the last gen Intel i9 MBP. Just visiting any non-basic website spins up the fans and battery life isn't great either. Yes, this thing is fast, but damn it gets hot just using it for normal tasks.

Still, I don't agree. I think this machine is meant to use local models. You just have to wear pants if you want to keep it directly on your lap. I rarely use it that way anyway. I prefer it plugged into an external display and comfortably sitting on a laptop stand.

My only problem with generative art is

1: right now, it's basically 90% beautiful hyper-realistic woman in some scene.

That is SO incredibly boring. These models are able to generate all kinds of never before seen mixed together content. Yet, everyone just wants to generate photos of things you could have already searched for.

2: creative people should be the biggest and loudest voices celebrating it but the vision researchers building the tools have only been interested in solving the math side of it and haven't bothered to ask creatives what they even wanted. It's fine in private R&D, at the hypothesis stage, but then these corporate heads went and shipped as finished products and charge us serious money for it. So, creatives rightly feel uneasy about the models that legitimately "stole" their content without permission, while also minting new millionaires at these companies by selling it back to them.

Once companies license the content and artists can make royalties, and the tools to interface with the models become sophisticated so that you're not one-shot prompting into existence a random thing, more creative people will come back. (i.e. ComfyUI)

---

Plus, these models need actual fully staffed product teams behind them to solve and own the ethical issues around things like character identity locking that can be used to deceive. Or being too easy to say "in the style of" that could almost perfectly replicate another artist's entire voice. Then, there is the nudity issue. I don't want generative models to have morals, especially when it's art. Art is meant to be provocative and boundary pushing. I realize that's an entire deep-hole of shit in there waiting to get through, but that's why we need people spending their work day thinking about it more than just blocking anything that shows too much skin.

And of course, gpt-image-2 has over-corrected and now as of today, prompts that worked fine a couple weeks ago are now getting blocked for "sexual content". I'm seriously placing rugby players on a field with poses, and the rugby play pose is "sexual" now. I don't want to see death, and I don't want to see nude people, but the censorship system is really horrible if any two humans touching each other in sport is now crossing the line.

Native software is incredibly difficult to build well.

There are at least 4 platforms they would need to support: Win, Mac, iPhone, and Android.

That's 4 different software engineers at least, just for the frontend.

Then, there's various backend engineers, who could be shared, yes, but not always. Android's weird runtime requirements are bespoke enough that just because the database is written in C++, doesn't mean it's the same C++ database as what the Windows backend would use.

Finally, there's the designers, who end up consolidating all the unique things about each native platform into a common design language so they can have a shared vision on all of the platforms. So engineers end up building UI that works identically on all 4 platforms, and you're basically building a bespoke "browser" at that point.

Turns out the browser is an incredibly sophisticated and highly performant layout engine that works on almost every platform out there. Native UI frameworks are always going to be more efficient and let you access more of the hardware, but it's more expensive to maintain 2 or 3 separate codebases.

It's a bit reductive to pin this on Ronald Reagan.

The entire western world had been shifting towards neoliberalism as a direct response to the eastern world shifting towards communism since WW2.

Trump also isn't the embodiment of anything other than the guy who didn't take it seriously and suddenly ended up with the job because the voters in the country decided it couldn't be any worse under him than whatever the current situation they were living with was.

Seems legit. My experiments with GLM-5.2 so far have resulted in strange hallucinations in the tiniest of places. Like a wrong variable name.

It seems like it's up for the task of complex code, but those little paper-cuts are scary to me. I wouldn't trust this model for anything remotely serious.

Vision decoding outside of the latent space of the model is lossy, but claude opus's vision isn't that great outside of UI screenshots. I mean it works in a pinch. At least in my testing, if you're looking at non UI images, there are better image to text models that can turn into a very precise documents that any LLM can easily parse.

I know at my company our leadership was quite indifferent to how AI was adopted and really relied on the ground level engineers to determine what AI was good for and how we can best make use of it.

It was the engineers at the ground floor who I watched become religious about it. They have been the ones pushing for deeper and deeper tooling. They have been the ones convincing leadership that this is the future and so now, well, it's leadership who is saying, we want more adoption.

This psychosis is happening at every level in our industry, and this isn't a big tech, or leadership problem. It's all our problems.

That billion dollars had to come from somewhere, and it definitely didn't happen from a single individual's hard work. A lot of people were involved in that over a long period of time. It's also weird to attach an emotional term like "earned" to that. You can say someone passively earned income, or an investor earned a return on their investment. That's clearly not the same meaning AOC implies when she uses it. So, yea. I'm pretty sure Graham and AOC are both talking past each other.

Can anyone help me understand why this particular issue is any different than Anthropic training its models with its brand of moral judgement since day one? I've always been turned off by their particular stances on things they bake into their models that steer users in directions.

Maybe this is just a different set of people now realizing that Anthropic does this and has always done this?

Do not forget that this company is launching this thing at the moment it's trying to IPO. It's not rocket science that their very public steering/denial claim is really just them hinting to interested investors that their moat is absolute.

Nano Banana Pro is still the industry standard as far as I’m concerned. I think giving a vision model spatial awareness is the next evolutionary step here, so I don’t think they’re behind at all.

Hollywood is using this tech already. Storyboarding and previs work has already fully become driven by AI tools.

The problem isn’t training data, it’s reference locking and allowing anyone to make whatever content they want.

I really do want to support artists, but I also feel super conflicted about what is actually at stake here if an AI agent generates a scene for me. I never would have hired a 3D artist before this moment, because there's no reason for me to. However, if I can easily poop out a 3D rendering of something custom without much time or cost, I would absolutely love to do that. How many one-off presentations or project design sessions I could have with cheap throwaway 3D artwork that provides value to explain my thought process?!

Just like AI image slop and AI book slop prove though, I highly doubt whatever Claude and Blender are cooking up will ever come close to taking a prompt like

render a scene of a corgi sitting on a chair looking out of a window at 3 cats playing with the corgi's favorite toy.

and turning that into anything useful.

GitHub had no reason to sell to Microsoft, they could have remained the bootstrapped company they started as, and rode the SaaS boom, since they were profitable on day 1. Seems a bit unfair to blame Microsoft though, because it was the founders who decided they wanted that sweet VC funding and Andreessen was happy to pay out.

Not sure if it mattered after that but they had that weird Tom Preston-Werner scandal that got him fired. Since he was the CTO, I kind of suspect that sent them on a collision course with needing to exit the VC round and Microsoft paid out.

I disagree. In college, I worked at a bagel shop, and one of my favorite parts of the job was washing the dishes and cleaning the equipment. It was incredibly therapeutic to have my headphones on, listening to music, and think about whatever, while also getting paid to clean.

I do use my dishwasher at home, and I love that dishwasher. However, I also cook and I want to get to bed at a reasonable hour.

Funny enough, I have come across many people in my life who usually only use their dishwashers as drying racks. It's a bit odd whenever I see it, but I get it.

So, people DO like drudgery. People even seek out drudgery. There is no one-size-fits-all of what it means to live a life, and as the article states, the biggest problem for adoption is that AI is so flattening as product.

I always understood, going to China, as what the industry was already doing, and Apple was in the middle of coming out of bankruptcy, so pressured to get their costs down. Tim Cook, the process guy, would have been given the task to do just that in an industry that was already consolidating in China.

I don't remember the details, but I'm pretty sure Tony Fadell's startup was already in China building what would become the iPod.

Tim Cook should be remembered, not for moving production to China, but for restructuring Apple's production lines to be built-on-demand, while also shipping those from China. It wasn't always perfect, and I bet the other people in similar roles and positions would have taken the easy path.

Let's see if Aaron Sokrin can make this compelling.

So, social media is your news source, same as it is today?

This is 10000% OpenAI's fault.

In 2022 the world was open arms, welcoming AI advancements.

However, since 2022, OpenAI and all of its original founding researchers, had their dramatic fallout, and began screaming in public saying crazy people things like "the end is coming."

Why did they insist on force launching ChatGPT? Google at the time refused to launch their own version (it was their own research that gave birth to LLMs) based chat because they knew all of the negative outcomes and unreliability of it all was just a poor product experience.

Instead of launch quietly like DALL-E and keep it fun and experiemental, nope, they threw it up online and moved full-steam ahead.

"THE END IS COMING" Sam Altman said. "AI WILL TAKE YOUR JOBS WITHIN 5 YEARS" Dario said. "AGI IS ALMOST HERE" Elon Musk said.

The disconnect is because these specific men, making those specific bold crazy person claims, with zealous cult following employees (including many of us here in this forum), kept marching ahead. Not only that, no one asked the rest of the world if they even wanted this technology EVERYWHERE.

This technology could have been so cool if it were given the breathing room to find usecases for it. Natural Language programming has been tried for a half a century, and it finally arrives.

Yet, it's so tainted by all the crazy person speak, and doomsday messaging, it's also thrown out there in such a haphazard way that have burned so many bridges, this technology is truely toxic. The fact that Gen-A and Gen-Z now have to waste brain power speculating if something is AI generated, is such a waste, but here we are. Welcome to the shit storm that was entirely made by those men.

Don't conflate businesses that have tech organizations in them, with actual technology companies. They are not the same thing. Technology companies exist as investment vehicles for R&D. There are endless streams of projects when your whole existence is structured around finding endless streams of projects.

No, this is 100% AI this time.

These companies need liquid cash to invest in AI infrastructure, and so they're doing mass layoffs to give them the capital to invest. Simple as that.

2022-2024 was COVID reset. 2025-2027? is reallocation of spending from R&D to infrastructure.

One broken limb or scratch would likely mean you're not going to live much longer. It would have been especially horrific if you broke your foot or leg and weren't able to do anything about it.

Also, children were easy food, and women definitely died in childbirth.

How many broken teeth did you just suffer with for the rest of your life? Scurvy wasn't just a disorder that was tied to long ship travel.

Mosquitos carried pathogens, the other food sources were also of random quality.

If you were born with poor eyesight or hearing, too bad.

I could keep going on, but the point is, being a self-aware mammal would have been absolutely torturous.

Can't imagine having to live with anxiety of just staying alive. Constant diseases, infestations, starvation, animal attacks.

You would never feel like you have time to just, be. Instead you're focused on getting your next meal, and finding a place to sleep.

It only took a few ice-ages to force us to get smart about how we organize and then here we are.

Neanderthals from 400kya are often classed as Heidelbergensis.

Heidelbergensis is the last common anscestor of Neanderthals, Denisovans and us.

We were all around for just as long, 400kya+, and before that, it was Homo Erectus.

All of them, Erectus, Heidelbergensis, Neanderthals, Denisovans and Sapiens were walking around at the same time. There's plenty of fossil records we've uncovered that show that to be true.

It was only in the last 100k years or so that we remained and the other variants "died out".

My company uses Claude through our own private data centers behind our own proxy that logs all requests and responses in and out. However, Anthropic heavily steers these models during RL to respond a certain way to certain prompting, so that's basically the "secret sauce" you're thinking of.

Or the more realistic, less tin-foil-hat reason, sophisticated chat bots are a real problem and dating platforms aren't immune to them.