Fully agree. I can drink coffee until 6pm and soft drinks until 9pm without impact to my sleep. There's little point in me moderating my caffeine intake.

Light, sleep time regularity, and exercise make a much more noticeable difference. That's where I should spend effort if I want to improve my sleep.

Even the official provider on OpenRouter seems to have this issue. Hope it's an easy fix for them.

Pay attention to what it forgets, and start telling it to proactively note down those things into nominated files while it works, e.g. indexing topics covered by each chapter/section/page of each paper/material. GPT-5.6 is really good at finding context again, if given the smallest hint where to start looking.

You can also ask it to read your past sessions, find places where it was wrong after compaction, and figure out a strategy to persist the most important summary information.

I have several Hermes threads that have each had >10 compactions with a 200k context limit, and with the right instructions on note-taking, they require at most a "double-check that against our past decision records" to be put back on track after a slip-up.

It's a solvable problem if you're willing to throw more tokens at it. Frontier models have gotten very good at cleaning up their own messes. You just need the right skills/loops, and to stick to models that consistently follow instructions (i.e. GPT-5.5/GPT-5.6-Sol).

There is a third case where the other party doesn't realize that the asker lacks the relevant experience to discern good LLM answers from bad answers for that topic.

Same solution as case one though - don't be afraid to say "Claude said X but that doesn't sound right".

Depending on the question that possibility can be quite rare these days. If you ask “How does x work in this codebase?”, it will read a bunch of files and give you a very likely to be accurate answer. If your using a platform without that context and ask it something more abstract, well, your mileage will vary.

Here's the problem, countering a lie or hallucination takes much more energy than asking Claude something and saying its true. Its the same as trying to fight misinformation on the internet, the amount of energy you have to spend to prove someone is lying or fabricating data is very high.

And having to do this on the corporate environment saps the energy and time of people that could be doing something productive by wasting their time answering a clueless person that asked an LLM about something they don't understand, got the answer they wanted (but that isn't real), and now are asking multiple people to prove it can't be done.

Here's an example, a PM decided they wanted to build a metrics framework, to track team success, with high level metrics. They asked claude to build such high level metrics (out of nowhere, these metrics don't exist), it happily produced hallucinated code that said it was collecting the metrics and the PR opened a pull request. Now we have to go there, review, find out is all bullshit and explain to the person that what they're trying to build doesn't exist.

So now we have to fight misinformation even on the clock.

Denying tech export to cooperative allies is certainly a move.

Many nations are now likely thinking: Why cooperate on international IP enforcement if we get lumped together with adversarial nations anyway?

As a current free-tier user, this price & transition plan seems completely reasonable. I'll upgrade when the nudge comes. I've got at least that much utility from it so far. Very glad this isn't a subscription.

I have found Claude to be especially unpredictable. I've mostly switched to GPT-5.4 now - although it's slightly less capable, it's massively more reliable.

Some of the flak is that issues are often only acknowledged once a fix is in place, and the partial fixes are presented as if they solve the whole problem.

The near-instant transition from "there is no problem" to "we already fixed the problem so stop complaining" is basically gaslighting. (Admittedly the second sentiment comes more from the community, but they get that attitude after taking the "we fixed all the problems" posts at face value.)

And they are often dismissed at first as perception/subjective bias, getting used to models being good and having higher expectations due to that, etc. users are blamed a lot before they are forced to admit that there is an actual problem.

Thank you for the reminder.

Childhood, at least IME, does a bad job of preparing people for this: fault and blame rarely matter in real interactions. When no teacher is around to play judge, all that matters is that you get a favorable outcome. That's something you need to figure out, usually in the heat of the moment, how to get. It takes active effort to understand the other side.