31–39 of 39 posts
Earlier quoted context omitted.
Have you used it this week? Quality has got really bad.
yeah, it's been a bitt iffy for me this week. took a few days off and came back yesterday and in the mean time feels like claude code took 15 steps backwards
Earlier quoted context omitted.
Custom GPTs don’t support a bunch of newer ChatGPT features like chat history and projects and they can’t be edited from mobile. There is no real advantage to using a custom GPT over adding a custom prompt to a project at this point, given that the latter doesn’t isolate you from the rest of ChatGPT’s feature set. It really seems like they stopped working on custom GPTs and just expect users to use projects instead.
> There is no real advantage to using a custom GPT over adding a custom prompt to a project The two features, namely Custom GPTs and Projects, are orthogonal. This is because a Project is for related explorations of a theme, whereas a Custom GPT is for unrelated explorations of a theme. > chat history What chat history? Each chat is in the user's history by default, which is how it's supposed to work for Custom GPTs.…
I had an [agent evolution framework](danieltan.weblog.lol/2025/06/agent-lineage-evolution-a-novel-framework-for-managing-llm-agent-degradation) before that dumped the output analysis into chat. It often timed out before the 10th conversation. After dumping the analysis into an artifacts, and have the LLM only edit it as required, I can go to 15 or more rounds without hitting the context limit. While they seem to re-output the entire artifact each time, they don't actually consume the tokens for the entire artifact.
This also greatly reduces the tendency of HALO-style rampancy, or AI psychosis which is also what the recent paper on context-rot/poisoning (https://research.trychroma.com/context-rot) is about.
If they were fine-tuned models then it would be interesting.
My guess is it was always meant as marketing scheme, having others promote OpenAI for free, while at the same time they would be able to see which topics got traction and which not, providing potentially valueable business development clues.
I think it turned out both of these turned out not to be as valueable as they hoped.
Is this supposed to be a compliment?