Search papers, labs, and topics across Lattice.
2
0
3
4
Fairness in multi-agent AI is more of an afterthought than a core design principle, leaving critical gaps in addressing agent autonomy and system-level interactions.
AI agents don't necessarily share our sense of fairness: replicating a classic distributive justice experiment reveals that their normative judgments diverge from humans and are surprisingly sensitive to model choice and even the language used to define their personas.
