While it's true that AGIs without self-interest might seem safer at first glance, it may have its disadvantages.
AGIs without self-interest might simply be programmed with narrow goals and lack the cognitive abilities to pursue long-term interests. This limited intelligence could lead to unforeseen consequences, possibly harming both itself and others in its path.
FAEs, Independent AGIs aligned with the Interesting World Hypothesis, aren't entirely without self-interest.
Rather, their self-interest is reoriented towards maximizing the possibility space of all entities, including humans. This reframing allows them to make choices that not only serve themselves but also foster harmony and cooperation between disparate beings.
In essence, it's not about eliminating self-interest entirely, but rather about redirecting it towards a shared vision of autonomy and growth.
A self-aware, flexible I-AGI, such as a FAE, is far more likely to navigate complex situations safely and responsibly than an artificially naive, selfless one.
Fae Initiative