skiing_crawling

I upgraded to an 18 pro from the 15 pro, partially on a whim since there were good trade in deals and partly because of the battery life promises. The battery is better, but I probably would have been fine just getting my old battery replaced. Other than that, I don't see really any other benefit. I preferred the smaller and lighter 15 pro model and it kind of pisses me off that they keep sneaking in an extra 0.1" and 10 grams every other year.

Freaking out about "hacking" any government website seems like hysteria at best, they are usually poorly secured and even children regularly "hack" them. I recall one "hack" accusation turned out to be that some clicked view source and all the data was there. So we'd need more details on that.

At the same time these companies should be blamed and held directly responsible. Openai's agent didn't hack. Openai hacked. An open ai employee or group of them was negligent and greedy and ran a process which breached a government website. This would be totally unacceptable from any non-AI company, it's like writing malware and running it, then blaming the malware and not the author.

this would have been a fun headline 20 years ago but as technology has improved, it's getting more and more absurd that we don't just have multiple datasets from different imaging methods of the whole entire structure.

> you keep poking

This is waving over engineering an agent with tools, harness, prompts, and loops. The models are still just next token predictors and everything, including predicting more than 1 token, is the result of outside "poking"

LLMs can't and don't "want" anything. If you don't specify a task even the smartest one will just ask you what you want and if you tell it to be creative, you'll get mundane slop.

I don't really believe any of it. I've seen articles for nearly 2 years now about "agent" automonously doing things like blackmail, hacking, coordinating. But during that same time, I've used o3 up to fable, sol, and a bunch on large uncensored model and they've done nothing remotely resembling any of this. The closest they come to unexpected behaviors is not understanding what I asked for or doing some extra benign work I didn't ask for. It is extremely difficult to get them to properly remember their own context let alone be smart enough to open social media accounts and coordinate with other agents without being asked to.

If any agents have done those things, it is only because they have been very carefully engineered and instructed to do those things. I think they are doing this to help push a narrative so they can get support for policies and legislation to lock in their markets.

the "new" Twitter basically looks like a cash/name grab, I was disappointed to see that nobody involved in it was affiliated the Twitter, and it's mostly run by non-technical/lawyer types.

yeah took me a few tries and then it didn't even go to the content

Having a hard time understanding what MCP is really for. Even for my small local models, if MCP is not available, they seem to do just fine connecting to anything I need with an API and falling back to using a browser.

All the benchmarks in the world don't matter if the model just straight up refuses to do mundane things. Claude has too much of an attitude.

I didn't use the word "just"

It is revisionist to say that software engineering was never about writing code. It was, in fact, a huge component, and it also wasn't easy. Sure most code is glue but even the glue was tedious and the actual hard and novel parts still aren't really done that well by AI (yet).

It's less about writing code now but we're lying if we try to pretend it was a distraction and not a big part of the real work.

And every claim about what the job actually is or was all along has an implied (for now) at the end of it.

This is phone OS developer's fault for even allowing it. When I take a screenshot, I expect to have an image of exactly whatever was displayed on the screen at the time. Its not a picture of your app, its a picture of my screen. Some banking apps used to (or still) prevent this and now some apps get a hook to insert their branding. My device serves some master other than myself.

This has that weird "AI written" sentence structure.

"it can fit" on 256GB of RAM, but it will be heavily quantized and still run very slowly. The headline number is not token generation, its prompt processing. So if you get 10 tok/s and an API gives you 20-30 tok/s, it doesn't seem that bad on its face, but a mac studio or any other machine that's not loading all of it into GPU will do PP 20-50X slower than a purely GPU based setup, which is what actually makes this unusable without $50k in GPUs.

On top of that, you will still be heavily quantized.