AI Risk Atlas Prototype/DemoUnofficial independent experiment. Not an official xAI product. Scores can be wrong.

Back to watch register
15Labor & rentsBelow the top 20

A workplace model that flags organising language

CapabilityDomain knowledgeImpact domainBoth

Statement (NASA form)

Given that mail scanners already score ‘risk’, there is a possibility of a legal activity treated as a threat resulting in speech at work as a feature.

Likelihood
3Probable
Consequence
4Critical
Urgency
3Priority

Personal slice · work. Experimental household card. Not a certified finding.

Composite 15 = 3×4 + 3

Applicable mitigations

Related on the map