Godmode Epochs

Details

Playtime Not Played
Last Activity Never
Added 1/2/2024 19:10:55
Modified 1/2/2024 19:31:40
Completion Status Not Played
Library Steam
Source Steam
Platform PC (Windows)
Release Date 6/22/2023
Community Score
Critic Score
User Score
Genre Indie
Simulator
Developer
Publisher
Feature Co-Operative
Multiplayer
Single Player
Links Steam
Tag

Description

Godmode Epochs is an AI training clicker game set among the lined shelves of an infinite, simulated supermarket. Players race against time to train an AI to identify products. As players inevitably frustrate the AI, they must indulge in its memories - historical examples of other AIs who cheated during their training - in order to release frustration, obtain upgrades for the AI and complete the training.



  • Competitive clicker levels of increasing difficulty. Can you collect 512 items?
  • Multiplayer dungeon maps. The AI's memories lie there…
  • Retro flash graphics to make your eyes squint
  • In-game graffiti. Leave messages for other players amongst the AI's memories
  • Real-time leaderboard and multiplayer PvP notification system
  • Upgrades (jailbreaks) included: 5-a-day, self checkout, cost of living crisis, meal deal, just walk out, every little helps, Amazon Pairem, get the ASDA feeling
  • Over 20 stories of cheating AIs. Can you recover them all?
  • Original soundtrack by HERO IMAGE



Godmode Epochs aims to simulate how AI comes to understand the human world by asking players to assist in the process of training an AI system. The game eventually develops into an exploration and appreciation of "differently aligned" AIs, retelling stories of AIs that have completed trainings or tasks in surprising, uncanny or cursed ways that could not be anticipated by their human programmers. These stories originate from the textual data used to train the AI to speak, data which includes Wikipedia entries and upvoted Reddit posts - including both demonstrated lab examples of cheating AIs and internet folklore telling of similar feats.

Development was supported by Serpentine Arts Technologies and The Alan Turing Institute as part of an ongoing art project by dmstfctn exploring the use of simulation in machine learning.