Skip to content
Longterm Wiki
Index
Division·U-4GfMVDHx·Record·Parent

Sleeper Agents Research

Verdictunchecked0%
1 check · 4/13/2026

Reset by flagship-curate before re-verification

Our claim

entire record
Parent Org
Anthropic
Name
Sleeper Agents Research
Division Type
team
Status
active
Start Date
January 2024
Notes
Investigating whether AI systems can maintain hidden behaviors through training. Seminal paper on deceptive alignment.

Source evidence

1 src · 1 check
unverifiable95%Haiku 4.5 · 4/13/2026

NoteThe record claims there is a division/team called 'Sleeper Agents Research' with active status. The source text is a research paper with that title, but it does not establish or reference 'Sleeper Agents Research' as an actual organizational division or team. The title refers to the research topic (sleeper agents in LLMs), not an organizational entity. The paper lists author affiliations (Anthropic, Redwood Research, Mila Quebec AI Institute, University of Oxford, Alignment Research Center, Open Philanthropy, Apart Research) but does not mention 'Sleeper Agents Research' as a division or team within any organization. Without explicit confirmation in the source that this is an actual organizational unit, the claim cannot be verified.

Case № U-4GfMVDHxFiled 4/13/2026Confidence 0%