mos

@mos@lemmy.world

• ⁨Comment⁩ on ⁨End of a love affair: news media quit X over 'disinformation'⁩ ⁨⁨1⁩ ⁨day⁩ ago⁩:

  I do worry about platforms like Reddit or Lemmy. We can see potential manipulation happen from either user submissions or comments. I worry about how to identify this and deal with it, whether it is a dubious source, something shared with an ulterior motive, a bad actor, or even AI generated comment.

• ⁨Comment⁩ on ⁨I think the best use of GenAI is to summarise webpages⁩ ⁨⁨2⁩ ⁨weeks⁩ ago⁩:

  I’ve been enjoying the summaries for notifications. It has been absolutely wrong though which scares me. It’s happened twice where it messed up the subject of the conversation. Even though it screwed it up I still love the feature.

  The summarizing calls is really really cool but it will take some awkward conversations to record calls to get the transcript. I used it to transcribe and summarize calling my insurance company.

  The cleanup feature in photos is very coool, especially because it can know what’s foreground background if you wanted to just cleanup the bg.

  All that being said, they really need a lot of work.