Conversation

Notices

1. Embed this notice
   Æ. (aesthr@wandering.shop)'s status on Wednesday, 26-Nov-2025 18:44:51 JST Æ.

   Last week in a seminar we discussed a text that was largely about sexual violence, including mass rape during war. Heavy stuff.

   One student admitted they had not read the text but worked off a ChatGPT summary.

   They had no idea the text was about sexual violence. ChatGPT withheld that information.

   This wasn’t just a minor error nor a typical LLM hallucination.

   About a third of the text, arguably its most important part, went completely ignored because it didn’t match OpenAI’s content policies

   • Haelwenn /элвэн/ :triskell: likes this.
   • Rich Felker repeated this.
   • Embed this notice
     Cassandra is only carbon now (xgranade@wandering.shop)'s status on Thursday, 27-Nov-2025 06:45:26 JST Cassandra is only carbon now

     in reply to

     @aesthr That's incredibly scary, not just directly, but also in the context of queer erasure, manufactured moral panics about CRT, and every other thing the right wing wants to censor out of existence. OpenAI has given them an immensely powerful tool for doing just that.

   • Embed this notice
     Markus 🎄 (markus@skvip.lol)'s status on Thursday, 27-Nov-2025 06:50:53 JST Markus 🎄

     in reply to

     @aesthr There was a similar story in Norway recently. One of our major telecommunications companies released a report on future threats, and war was the major one.

     A journalist used an LLM to summarise it for an article, and no mention of war came through. Only the lesser threats.