I'm using https://flag.helios.surf/ to issue flags. I plan on delegating to @helios.voter so I can earn some extra tokens to burn for summoning the downvoter. I think asking for hp delegations is an excellent idea, I might need to create an incentive token economy to get that going through. I will look into renting a few hive delegations too if I can get a good deal.

As for the AI-Generated text detector, I've been having more luck with GPT 3.5. I've noticed some discrepancies between different detectors, but mainly that happens when the detector only detects parts being written with ai. For the most part, these detectors do seem to get the job done, haven't noticed any false positives when testing my own original content in them. Nevertheless, I'm still going to have to be very careful and take into account the big picture too, the history of the user is just as important when relying on the detector, investigating for things like changes in writing style and comparing works from pre-gpt and post-gpt writing samples if possible. Also using one's own judgment to determine the difference between ai abuse with intentions to mislead readers, vs simply paraphrasing their own original content ideas.

Just like everything, these detectors are not a perfect science, so I'm going to be very careful and take this work one post at a time.