New in Nat Mach Intell: interoception—monitoring and regulating internal bodily states—may be the missing ingredient for AI agents that stay autonomous and adaptive over long time horizons as environments change.
doi.org/10.1038/s422...
#AI #Neuroscience with @sungwoo.bsky.social and COMBINE lab
doi.org
Life-inspired interoceptive artificial intelligence for autonomous and adaptive agents - Nature Machine Intelligence
Lee, Oh et al. propose interoception as a biologically inspired framework for building more autonomous and adaptive AI agents, learning from living organisms to build autonomous and adaptive intellige...
![There is a “centre” where the money, the fame, is; most likely your proposal gets funded because it’s on the most favourable topic. Maybe today, RNA is [most favourable]. If you are working with mRNA, maybe that's the centre there.
And then there are people in the periphery. There is no fame, there is no money, no nothing there. The only thing in the periphery is freedom. You can do what you like to do, what you feel is important.
Here’s what a proposal is: why they should give me money. And they should question that. “She came from university nobody knew about.” “She never had a mentor who was famous.”
And somehow it gravitates always to the same people, same circle. They get published there, they get the money. And that's another explanation: I was not famous enough or didn't have anybody who would support me in a way that somebody that’s a famous and well-established scientist stands behind you and says, “Oh, look at this, it’s good.”](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:h6srr75ahvugj6gts7ql4cpl/bafkreie7jkxonqwgbbcuni4dh54nhodteq24wv4dp6smnvsomdpwpve4we)
