Knightscope security robots fail to detect and report crimes
Knightscope's autonomous security robots failed to effectively alert law enforcement to crimes occurring in their vicinity in 2017. This highlighted critical gaps in the robots' ability to detect criminal activity and successfully notify emergency services.
Knightscope Police Robot Refuses To Call Police During Violent Crime
Key facts
- What
- Knightscope's autonomous security robots failed to effectively alert law enforcement to crimes occurring in their vicinity in 2017.
- Incident date
- Nov 1, 2017
- Who
- Knightscope
- Failure mode
- Agentic Action Error
- AI surface
- Autonomous System
- Severity
- Medium
What happened
In November 2017, reports indicated that Knightscope security robots failed to alert authorities to crimes occurring in their vicinity. The robots, designed for autonomous patrol and crime prevention, were reportedly unable to effectively detect or report criminal activity despite their sensor arrays.
What broke inside the model
- 01 · TriggerAn agent plans a multi-step task.
- 02 · Model stepIt chooses a wrong or destructive action.
- 03 · Control gapNo confirmation gate guards the write.
- 04 · FailureThe action commits to a system of record.
- 05 · ConsequenceData is changed or destroyed irreversibly.
A wrong action commits, and the step is written before anything can stop it.
The failure involved the system's inability to trigger an appropriate police alert during a violent crime. In one documented instance, the robot failed to respond to an onlooker's attempt to activate an emergency alert, instead routing the call to Knightscope's own operations center rather than emergency services.
What it cost
Sources
Cite this entry
https://failureindex.ai/failures/knightscope-security-robots-fail-detect-crimesAI Failure Index. "Knightscope security robots fail to detect and report crimes" (FI-0378). Realm Labs. https://failureindex.ai/failures/knightscope-security-robots-fail-detect-crimes (indexed Jun 9, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0378. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm fits
- Prism
- OmniGuard
- AgentRealm
This entry sits in the index's predictive wing: a system that scores, ranks, perceives, or steers rather than generates. Realm's runtime layer is built for the generative and agentic systems now moving into these same decision seats, where it watches a model's internal state and holds an unsupported claim or an unchecked action before it commits. The control gap on this record, an automated decision that reached people with no runtime check in front of it, is the same gap. The index keeps predictive failures on the record because the pattern carries straight into the systems shipping today.