{"ID":2845243,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.04177","arxiv_id":"2511.04177","title":"When Empowerment Disempowers","abstract":"Empowerment, a measure of an agent's ability to control its environment, has been proposed as a universal goal-agnostic objective for motivating assistive behavior in AI agents. While multi-human settings like homes and hospitals are promising for AI assistance, prior work on empowerment-based assistance assumes that the agent assists one human in isolation. We introduce an open source multi-human gridworld test suite Disempower-Grid. Using Disempower-Grid, we empirically show that assistive RL agents optimizing for one human's empowerment can significantly reduce another human's environmental influence and rewards - a phenomenon we formalize as disempowerment. We characterize when disempowerment occurs in these environments and show that joint empowerment mitigates disempowerment at the cost of the user's reward. Our work reveals a broader challenge for the AI alignment community: goal-agnostic objectives that seem aligned in single-agent settings can become misaligned in multi-agent contexts.","short_abstract":"Empowerment, a measure of an agent's ability to control its environment, has been proposed as a universal goal-agnostic objective for motivating assistive behavior in AI agents. While multi-human settings like homes and hospitals are promising for AI assistance, prior work on empowerment-based assistance assumes that t...","url_abs":"https://arxiv.org/abs/2511.04177","url_pdf":"https://arxiv.org/pdf/2511.04177v1","authors":"[\"Claire Yang\",\"Maya Cakmak\",\"Max Kleiman-Weiner\"]","published":"2025-11-06T08:29:29Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.MA\"]","methods":"[]","has_code":false}