Yep, Rules Engines and Expert Systems are pretty nice in some ways.

I guess one might get away with calling your startup an "AI" startup now, and just be quietly using classic AI actually-reliable Expert Systems instead of the current hyped ocean-boiling LLM statistical bullshit-spewing.

CLIPS is forward-chaining, whereas Prolog is backward-chaining.

https://en.wikipedia.org/wiki/Rete_algorithm - typically used to implement forward-chaining (including CLIPS) and is kinda neat.

https://en.wikipedia.org/wiki/SLD_resolution#SLD_resolution_strategies - Prolog of course does depth-first-search with backtracking as its builtin strategy.

The widely-used Drools Business Rules Engine is another one you might encounter. It does both forward and backward chaining. https://www.drools.org/

AI-Winter effects means stuff like this is not always called AI. But it totally is classical "AI" subfield stuff.

https://docs.drools.org/6.0.0.CR1/drools-expert-docs/html/ch01.html#d0e29

Over the last few decades artificial intelligence (AI) became an unpopular term, with the well-known "AI Winter". There were large boasts from scientists and engineers looking for funding, which never lived up to expectations, resulting in many failed projects. Thinking Machines Corporation and the 5th Generation Computer (5GP) project probably exemplify best the problems at the time.