Co-evolution of self-replication and function in a digital primordial soup

Re:Co-evolution of self-replication and function in a digita... by vatsachak on Sunday July 19, 2026 @01:38AM (#48965287)

Interesting. But, evolution is way too unconstrained to provide us a path to "agi". It would require too much compute.

Evolution also eventually gets frustrated and creates the brain, capable of in context learning.

Maybe we should take some notes from these massively parallel, shallow, and highly recurrent constructions.

[ Parent | on HN ]

Re:Co-evolution of self-replication and function in a digita... by bob1029 on Sunday July 19, 2026 @03:36AM (#48965758)

If evolution had access to TSMC's process technology, do you think it wouldn't leverage it? The signals in my AMD CPU propagate ~1,000,000x faster than the ones in my brain.

[ Parent | on HN ]

Re:Co-evolution of self-replication and function in a digita... by eyvindn on Sunday July 19, 2026 @06:56AM (#48966843)

^ yes. this is what we hope.

[ Parent | on HN ]

Re:Co-evolution of self-replication and function in a digita... by vatsachak on Sunday July 19, 2026 @12:00PM (#48969286)

Super big fan of your work on cellular automata. Do you have anything interesting papers coming up?

[ Parent | on HN ]

Re:Co-evolution of self-replication and function in a digita... by eyvindn on Tuesday July 21, 2026 @06:44AM (#48990528)

yes! very soon!

[ Parent | on HN ]

Re:Co-evolution of self-replication and function in a digita... by vatsachak on Sunday July 19, 2026 @12:02PM (#48969294)

Yeah. But we need throw money at studying the brain, instead of another billion to "agents which understand you truly"

[ Parent | on HN ]

Re:Co-evolution of self-replication and function in a digita... by tim333 on Sunday July 19, 2026 @09:02AM (#48967871)

Real evolution produced human intelligence.

Constrained evolutionary algorithms may have some promise.

[ Parent | on HN ]