The limits are what did it for me. They kept boasting about Opus performance and improvements, practically begging me to try it out, and when I did, it totally obliterated my usage. I'm sure its good, but I stick to Sonnet because I've been burned bad. Never had that problem with ChatGPT, but it turns out they're just unprincipled and evil, which is a shame.

Good. Proud of her. We need more like her who have principles.

What a cool idea. Love it!

Yeah this is a huge problem for Anthropic and its why I unfortunately haven't been able to kill of my ChatGPT subscription yet.

Me either. I guess it's just fatigue, at least for me. I also don't really get that excited by new LLM releases either. Not to say the tech isn't impressive, but I guess all the hype has me inured.

Sounds pretty beefy. What kind of local LLM is that thing capable of running? Does it open up real alternatives to cloud providers like OpenAI and Claude, or are the local models this hardware is capable of running still pretty far behind?

Deleting my account today once I import my data to Claude

I'm also waiting on my ChatGPT data export. I started it last night and I'm still waiting. I would say there's huge opportunity here for Claude to offer direct import tooling.

Literally a feature being advertised as of today.

Let me also register my disinterest in another platform that mimics TikTok.

Brainrot is brainrot regardless of whether it's "federated and open-source" or not. And BTW, 95% of the people on TikTok don't care about either of those two things.

Good write up I guess, but I'm just so tired of all the AI-isms in every damn thing.

"Your European passport is one quiet subpoena away"

Why does the subpoena need to be quiet? If I search my chats with ChatGPT for the word "quiet", I get a ridiculous number of results. "Quietly this, quietly that". It's almost like the new em dash.

There's many others all over this blog post I won't bother calling out.

"Understanding what I actually agreed to took me an entire weekend reading 34 pages of legal documents."

Yeah I'll bet it did. Or it took an hour of back and forth with ChatGPT loaded up with those 34 pages.

I get it, we all use AI, but I'm just so tired of seeing the unmistakable mark of AI language all over every single thing. For some reason it just makes me think "this person is lazy". The CEO of a company my friend works for used Claude to write an important letter to business partners recently and we were all galled at her lack of awareness of how AI-sloppified the thing was. I guess people just don't care anymore.

I also find AI trope-ification articles exhausting to read, there's a reason I've fine tuned my system prompts to wipe all of it away. This reads like "Hey Gemini, I verified my passport on LinkedIn, write an impassioned exposé on Persona's privacy policy".

When people leave in things like staccato language and Blogspot era emphasis, I feel like I might as well copy the Persona privacy policy and prompt my own AI(s) on the topic and read that instead.

> Or it took an hour of back and forth with ChatGPT loaded up with those 34 pages.

That's exactly what I was thinking when I read that line. And there's nothing necessarily wrong with using AI to help decipher large legal documents, just be honest about it.

Or just verify and write up its findings yourself, this is like pasting notes from a research assistant in verbatim. It comes across as pretty lazy!

The part of the story I believe is the part about basically half naked children on Facebook, whether real or AI. I haven't put anything on my profile for the algorithm to tailor content to, since it's used for only marketplace and I've seen some very disturbing content that looked like it slithered off of X. It was as suggestive and inappropriate as you could be about kids, without being full-on porn. And Facebook/Meta seem to have no problem with it. It's a trash heap of a site and everyone involved with it working at Meta should be ashamed.