Aligning AI Teams
Siddarth Srinivasan, Morgan Jane Matthews, Jascha Sohl-Dickstein, Erik Jones
Aligning AI Teams: 8 upvotes on Hugging Face Daily Papers, #20 of 40 papers on 2026-11-16. Day-by-day upvote history.
Upvotes over time
It had 12 upvotes by its first day, 11 upvotes by the next day, 8 upvotes by day 3, 8 upvotes by day 7, and has 8 now. It ranked #56 of 146 papers in its week and #264 of 520 in its month.
Previous work has shown that alignment training on individual LLM agents can fail to transfer to multi-agent settings. We investigate the underlying mechanisms for this failure, and explore interventions to preserve alignment transfer to multi-agent LLM teams. We focus on two software-engineering tasks (sepsis triage, a news recommender) and ten consulting-proposal generation tasks. We present four findings: (1) The single-vs-team safety gap does not show a clear trend with more capable models. We find that Anthropic's Mythos Preview shows the largest gap on one of our tasks but the smallest on another; (2) Many natural interventions, such as having agents critique each other's work or coordinate in a group chat, do not reliably bring agent teams back to single-agent alignment behavior; (3) diffusion of responsibility substantially accounts for the safety gap: agents in teams are less likely to proactively check for system-level safety failures and more likely to ignore, rationalize, or deflect responsibility for issues when they arise; (4) having a designated safety lead agent ensure system-level safety, and having a coordinator agent publish a safety specifications document are effective at restoring single-agent alignment. We also release magelab, a multi-agent LLM orchestration and experimentation framework. Our findings can help practitioners design agent teams that successfully preserve the behaviors expected of aligned individual agents.
Paper page on Hugging Face · arXiv
Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.