Wow, I'd totally forgotten that this execellent paper was about AI! When I last read it, AI was more or less dead, at least under the 'AI' label, so I didn't pay attention to it. Now that AI is undead, it's going to be a different kind of read!
Comments
Log in with your Bluesky account to leave a comment
The symbolic logic AI in Agre's writing differs in many technical ways with 2020s deep learning, so it's surprising how on-the-mark the critique still reads. There's still the same deliberate vagueness in how the world gets distilled into symbols and representations to begin with.
But Agre's points aren't always limited to AI. I riff a bit on this in the context of mappings for DMIs in my recent JNMR paper -- §3.1, which also draws on some work by @thormagnusson.bsky.social and @adamlinson.bsky.social. https://www.tandfonline.com/doi/pdf/10.1080/09298215.2024.2442356
Comments
https://orpheusinstituut.be/en/publications/sound-work