On Tue, Oct 27, 2026 at 6:03 AM Martin Steigerwald <[email protected]> wrote:
...snip... > Now I looked at an assistant report more closely and I did not see a hint > on which model was used for the review[6]. But what has really taken me > aback was this: > > "Uploader's response to the LLM review > > The uploader was given 24 hours to respond to the serious issues the > review found, and did not." > > I am not kidding you. That is really written out on the public assistant > report webpage like that. > > The uploader is presented with a wall of text and then they were given 24 > hours to respond to it? > > Honestly I am without words about this. > > This is not okay. Simple as that. It's entirely possible the chatbot is just doing that without being directed to, or that it was told to do something along those lines but for reasons other than forcing people to reply in a very short time. In Kicksecure, we use quite a bit of AI stuff for code review and prototyping of things that are later thoroughly reviewed and polished by a human, and quite frequently Claude will look at a change I've made and say something along the lines of "Aaron has undone (insert whatever asinine change it made that I had to unbreak here). Remove (the working code I put in its place) and restore (our asinine change)." Of course, neither me nor the developer doing the prompting has any desire for my fixups to be destroyed in favor of unaltered vibed slop, but the bot does that anyway, and we just ignore it. Unless someone is trying to enforce these kinds of requirements, I wouldn't read too much into it. More than likely whoever reads that when looking for something actionable to do will chuckle a bit at the model's overzealousness and then ignore it. -- Aaron
