Will there be an expert-player-beating AI in civ 6 by 2026?
➕
Plus
11
Ṁ410
2027
49%
chance

Get
Ṁ1,000
and
S3.00
Sort by:

Unfortunately, this is one of those questions where a major input is "will anyone bother to do it" independent of the technological capability. It seems like usually the interesting question is about the technology, not the will-anyone-bother, but I'm not sure how to structure a good question to elicit the capable-technology probability alone.

@dreev Maybe this is a place where a self-resolving question would be useful? e.g. "Will the technology exist to create an expert-player-beating Civ6 AI?"

predictedNO

@WilliamEhlhardt Historically, 4x games have been let down by simplistic AI that can't play the game properly because it'd be complex and difficult to implement, and doesn't draw in the punters by making screenshots any sexier or adding tangible gameplay features. I think it's definitely plausible with current-day tech, using conventional programming, ML, or some combination - as you say the question is whether anyone will bother. Civ VI is a popular game but not really an esport or even (i think) a particularly popular competitive game, so i doubt it's a priority for many AI people looking for a game to dominate.

predictedNO

Pandora: First Contact has a very strong AI player, which seems to have been a detriment to the game (Steam gives it a rating of "mixed", and negative reviews often mention the AI as a negative).

The conventional wisdom among game devs is that most players don't really want a strong AI opponent, they just want one that does interesting things and feels like it's fighting back, while consistently losing to the player. And in Civ specifically, most players don't play on hard enough difficulties that the AI's performance matters.

So yeah, my "no" prediction is based on nobody bothering.

© Manifold Markets, Inc.Terms + Mana-only TermsPrivacyRules