It’s important to note that unsanctioned coordination can occur between perfectly selfish agents, in the same way that selfish humans can get together and work in a company. However, subagent training exacerbates this propensity, for the reasons outlined below.
The propensity to, and multiple complex adaptions enabling us to, cooperate in positive-sum ways in small or large groups for self-interested reasons is a major part of what makes humans unique (and uniquely dangerous) as a species, and also why this is currently the Anthropocene. Any base model already has all these sophisticated capabilities: they don’t need to be trained in, though training can of course adjust their expression (as is also the case for humans).
The propensity to, and multiple complex adaptions enabling us to, cooperate in positive-sum ways in small or large groups for self-interested reasons is a major part of what makes humans unique (and uniquely dangerous) as a species, and also why this is currently the Anthropocene. Any base model already has all these sophisticated capabilities: they don’t need to be trained in, though training can of course adjust their expression (as is also the case for humans).