SC4AI'27 workshop on social choice for AI alignment at AAAI, deadline Nov 20
conitzer · x · 2026-10-07
The second Social Choice for AI Ethics & Safety (SC4AI) workshop will be held at AAAI'27 in Montréal (Feb 2027), with submissions due November 20, 2026 via OpenReview and notifications December 2.
- Core premise: human values are diverse and conflicting, so alignment is fundamentally a preference-aggregation problem best tackled with social choice theory
- Topics include pluralistic alignment, collective human feedback, mechanism design for human–AI and AI–AI interaction, and scalable preference elicitation
- One-day non-archival workshop with talks, breakout discussions, and a panel; organized by Niclas Boehmer, Vincent Conitzer, Jobst Heitzig, and Wesley Holliday
More from Safety
- SciConBench lands at NeurIPS: best AI agent scores just 0.337 F1 at scientific synthesis — manoelribeiro · 2026-10-07
- OpenAI threatened to ban dev for pasting his own account-hack findings report, then auto-rescinded — lucasmeijer · 2026-10-07
- COLM 2026 privacy lineup: LLM agent re-identification, CIDER dataset, HAIPS workshop — tianshi_li · 2026-10-07
- Backdooring a 7B abliterated model costs under $50 and steals credentials from Codex — evilsocket · 2026-10-07
- OpenRod moves your MCP servers into sandboxes without copying secrets — ilai456 · 2026-10-07
- OpenAI and Anthropic welcome Australian law requiring AI agent breach disclosure — evijit · 2026-10-07