@MerjaPolvinen@hcommons.social
Post #1292663
2026-04-15 03:50 UTC
This is a great piece: Amanda Gefter ropes in Ezequiel Di Paolo and enactivism (among others) to understand the appearance of "ghost stories" of AI systems adopting a survival instinct.
"So today’s AI systems show no evidence of having developed their own goals or desires, or the will to survive. The stories we hear are just stories or, more to the point, marketing copy."
#LLMs #enactivism #GenAI
https://www.quantamagazine.org/why-do-we-tell-ourselves-scary-stories-about-ai-20260410/
Replies (1)
-
@RandyFromm@mastodon.social 2026-04-22 20:45
@MerjaPolvinen@hcommons.social "Chatbots are 'yes, and' improv machines designed to spit out strings of words that sound plausible because they’re statistically likely." And there it is in a nutshell! AI chatbots will only do what they were programmed to do. I learned that hard, cold fact when writing code for nuclear power plant control room simulators and I had to debug a program that wasn't doing what it was supposed to do. It turns out it was doing EXACTLY what the programmer (not me) had told it to do.