timfsu

I'm not an Anthropic fan, but it's worth asking - why is it always OpenAI?

Subscriptions are, and will likely remain, the best deal in town. Unfortunately, larger companies aren't able to do that. When your monthly token costs are in the $5-10k range, the local inference starts to look a lot more attractive

Fascinating article. I daily catch LLMs in “lies” like: “I found the root cause of the bug” or “this approach is twice as fast”. It’s hard to say what causes this uninformed certainty - is it intrinsic to being trained on human writing, or something that comes from the RLHF process afterwards, but it’s extremely annoying. It’s one thing to have a LLM make poor decisions, but it feels worse to have it “lie” to you in the process.

More accurate to say restaurants are about people, not food. May be an apt analogy - chefs and cooks spend their time in the kitchen preparing food for a diner they don’t see.

This makes me think - should we be using non-monospace fonts for coding agent chat to improve readability? Wonder if we can get Claude and Codex to do that automatically in supported terminals

In my testing, models don't really know their own name - I would be very suspicious of --model actually revealing the real model name

I had exactly this use case - in a grocery store in the Alps, no internet, fired up a local LLM on my phone to figure out what to cook and what to buy

Wow, this is pretty scary. LLMs have made phishing attempts look so much more legit, and the damage they can do so much greater.

This is big, but until we have policy clarity we can’t trust it. I’ve always migrated all of our agents to Pi SDK, we aren’t going back

In contrast, I’m on the $200 max plan for Codex and I hit the 5 hour limit near daily at work. I typically am having it work on about 5 tasks an hour. I’ve never hit a 5 hour limit on Claude on the $200 plan but I have hit my weekly limit.

Yeah it’s hard to call that cheating from a model. Maybe “disqualifying” is more accurate

We saw this too with Gemini specifically. My favorite example - we built a hallucination detector (given the input, does the output make any false claims) in Gemini, and after the Seahawks won the Superbowl in February, it would consistently flag that as "not possible".

Imagine you try two products you’ve never heard of. You prefer one over the other. Was it marketing? That’s what’s happening here. Marketing can get you to try something you wouldn’t have otherwise, and it may suggest benefits you’d get if you tried it, but your preference of using one thing or the other is a subjective experience of your own.

This is dope! We basically built something very similar internally for our team and it's been a very natural and intuitive way to manage agents (as opposed to having a bunch of terminals to track). Not every task/conversation can be done in the background, so it's been helpful for us internally to be able to seamlessly transition between "interactive conversation" and "background job done by agents" even within a single card.

Yes, but they expire in ways that unions don’t