I’m starting to think of ChatGPT as more dangerous for kids than many of the hard drugs out there. Honestly starting to think it should be outright dismantled as a criminal enterprise.

Are you kidding me? Have you seen the state of software? Have you installed the vibe-coded Windows patches? Looked at the horrific slop overflowing on GitHub?

LLMs are fast as coding assistants. Fast doesn’t mean good. The drop in quality over the last year or so has been nothing short of shocking. And I’ve had multiple vibe-coders tell me my mistake is looking at the code, which they no longer do.

Let’s not. I’d like you to tell me what I got wrong. I described the output of LLMs as “a bunch of pablum [which means ‘trite, insipid, or simplistic writing or speech’] that sounds convincing and sycophantic[.]”

So do you want studies that show they produce pablum, studies that show their output sounds plausible and convincing even when it’s wrong, or studies that show they’re trained for sycophancy?

What I posted was correct.

Because I’m capable of reading chat logs and research posted by humans. I follow this stuff closely, because I’m a software developer and it impacts my career. I don’t have to use LLMs to understand how they work and what they do.

I’m considering moving much of my 401K to the emerging markets equity fund. Not much exposure to AI there.

I’m… not? I don’t touch the stuff.

You’re right, it responds with a bunch of pablum that sounds convincing and sycophantic, too.

Oh boy! Another app that might randomly delete important files or send fake tips to police departments because it’s driven by a statistical token generator! I can’t wait to never install it! Yippee!

If we ever manage to get the country back, we need to immediately ban AI-created political advertisements, and put teeth into the law.

Boom Bubble. It’s not a boom, it’s an economic bubble, propped up with circular financing instigated by NVIDIA.

Because how else will they enshittify it?

Sure, but how likely is that?

I’m 100% positive they’re not reading my ChatGPT logs.

(I don’t use ChatGPT, and neither should anyone else.)

No. We’re starting to figure out that OpenAI is both incompetent and unethical. Nothing is “escaping.” LLMs are just burning compute to accomplish tasks they’re assigned by incompetent and unethical people at the company.

If they set up their sandboxes properly, none of these hacking incidents would happen.

Let’s make one thing clear: The “extinction warning” was marketing.

LLMs generate text. That’s all they do. Don’t want them to create an army of killer robots or deadly viruses? Then don’t run them in a harness that can create robots or viruses. Problem solved.

I think you’re misunderstanding what’s happening. The LG TV isn’t directly using the Windows internet connection. It’s taking advantage of Windows Update to have Windows itself install malware as a driver update for the screen.

Which I think is actually worse, because it shows that Microsoft is complicit in this nonsense.

It’s already happening. Software instability is becoming a major problem.

Honest answer? Not much… yet. As the stupidity of “tokenmaxxing” and replacing developers with hallucinations-and-plagiarism engines really comes into focus, I expect developer hiring to ramp back up.

Meanwhile, in the real world, my company has locked down usage of Anthropic’s top-tier models, because costs were out of control. Even the low-end models are usage-capped, and developers are starting to write code by hand again.

And we’re busily setting up local models we control ourselves on our network edge. We’ll probably end up with a modest budget for frontier models (until the model providers collapse), but most of the coding will be done by local models and developers themselves.

And that seems to be the way the entire industry is going, unless Anthropic starts subsidizing tokens with investor money again.

In conclusion: Get fucked, Dario.