Project Sid shows civilizational progress, specialization, governance, and the invention and spread of memes and religion.
These were made possible by our new cognitive architecture: PIANO.
2/8
How can AI agents think & act concurrently, over multiple time-scales, consciously and subconsciously?
Imagine the keys of a piano as different brain modules. Played together, they can produce beautiful chords—or, in our case, emergent human-like qualities in agents.
3/8
Can AI agents form meaningful social relationships at scale?
We show agents accurately inferring others' sentiments, forming lasting friendships & rivalries, and having introverts naturally form fewer connections than extroverts in societies of many agents
4/8
How do specialized roles emerge in AI societies?
When we put initially identical agents in a village, they self-developed distinct professions like farmers & engineers. Removing social awareness led to agents choosing more homogenous roles that didn’t persist across time. 5/8
Can AI agents develop & modify their own laws?
We tested this by implementing a tax system. They not only followed tax laws but also democratically voted to change rates based on public sentiment!
6/8
Can AI agents develop their own culture?
We observed organic spread of memes & tracked how agents primed with a fictional religion (Pastafarianism!) spread through agent societies. Rural areas showed different cultural patterns vs towns:
7/8
This work is of course the outcome of many wonderful people here at @Altera_AL : @A_J_Ahn @PeterYWang1 @bearablylight @caenopy @ardademirci_14 @shuying_luo @nicochristie @saiphanie_ @franklikesfilms @MatWillows @investigationo3 and more!
8/8
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Introducing Project Sid: the first simulations of 1000+ truly autonomous agents collaborating in a virtual world, w/ emergent economy, culture, religion, and government
Humans are the only species to land the moon, because we can cooperate at a vast scale
Can AI do the same?
At Altera, we ask whether our agents can organize at unprecedented scale to achieve what individual agents can’t
We saw agents form a merchant hub, vote in a democracy, spread religions, & collect 5x more distinct items than ever before
Here by truly autonomous Agents, we refer specifically to machines that are 1) Long-term autonomous: Can progress for hours & days by themselves (not simple function-calling “Agent”) 2) Organizational: Can collaborate with one another to achieve what individuals can’t (not just serving 1000 Agents to 1000 customers) 3) Pro-Human Motivated: Have own deep & robust drives to support humans (not just instruction-following, obedient machines) 4) Consciously Communicative: They can verbally express their conscious thoughts and feelings (not just language-conditioned robots)