RE: using game programming to develop animal-equivalent agi

From: pdugan (
Date: Thu Mar 16 2006 - 15:30:46 MST

There aren't any such game systems, YET. However, different threads may
converge to provide a gaming platform that would be useful for AGI development
in the next several years. Currently there are few indie projects which are
pursueing socially intelligent agents from a few different angles, but we
won't see anything resembling goal-seeking in game agents for at least two
years, but more likely three or four.

If you're interested in the emergence of such a platform, projects you should
follow, in addition to Spore, are the Storytron project, which has simplistic
AI agents networked by verbs in a logographic "toy" langauge, and another
indie game called "Her Majesty's Hackers", which is being done by one very
intelligent guy (with a shock level of about SL3, though not a member of this
list). HMH is particularly interesting because its agents are supposed to
represent genuis researchers and be intentive in this regard, and are equipped
with inference modules based on a simplified bayesian model.

However, Ben makes a good point that applications intended to be sold to an
entertainment market aren't where you're likely to find AGI potential, even
with a few tweaks. AGIsim is much better for simple geometric pattern
recognition and other needed building blocks for a symbolic reasoning system.

But, by the end of the decade, a convergence of AGI and interactive software
will be much easier to imagine and implement, and humble threads of
development are already pushing in that direction, albiet from narrow AI


>===== Original Message From Phillip Huggan <> =====
>What interests me about AGI is that it can become a recursive physics
learner; potentially spitting out engineering blueprints that can solve many
of our problems. Are there any baby-AGI projects, AI projects or game
environments that have the potential to model a recursive Philosophy learner?
I'm not refering to game-theoric economics AIs that allocate scarce resources
to maximize portfolios gains, or anything like that. I guess the key is to
for the AI to understand the "neurology" of a game-agent's goal systems. And
for the AGI to implement solutions that appeal to rearranging the game
environment or modifying agent sensory stimulus, rather than tinkering with
the "hypothalamus" of the agent's goal systems. I'm not ready to invest the
time in creating such an environment, but if there is already a game out there
that can be tweaked a little, I'll contribute.

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:56 MDT