Abductive Inference, and more specifically "Parsimonious Covering Theory" (PCT)[1] which is one approach to automating abduction. It's a not a new idea by any means, but I think there's still some meat on that bone (so to speak). I've been working on re-implementing PCT using in a more modern environment, with an eye towards exploring ways to extend the idea and/or integrate it with some more recent approaches.<p>[1]: <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2244953/" rel="nofollow">https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2244953/</a>
<i>Kingdom of the Wicked</i>, a novel set in a world in which the ancient Romans experience an Industrial Revolution and have approximately contemporary levels of technology when Jesus shows up: <a href="https://jakeseliger.com/2017/12/28/kingdom-of-the-wicked-book-one-rules-helen-dale" rel="nofollow">https://jakeseliger.com/2017/12/28/kingdom-of-the-wicked-boo...</a> . It sounded gimmicky to me but is shockingly well done.
Directed Acyclic Graphs / block lattices in Cryptocurrency, like those used in IOTA and RaiBlocks:
<a href="https://raiblocks.net/media/RaiBlocks_Whitepaper__English.pdf" rel="nofollow">https://raiblocks.net/media/RaiBlocks_Whitepaper__English.pd...</a>