AI firm Evolv accused of exaggerating AI weapon detection claims

The rapid adoption of AI-powered security technology in schools and public spaces has come under scrutiny as regulators crack down on companies making unsubstantiated claims about their capabilities.

Critical findings: The Federal Trade Commission has determined that Evolv Technologies made misleading claims about its AI-powered security scanners’ ability to detect weapons while ignoring harmless items.

• The company’s scanners, deployed since 2021 across venues like schools, sporting events, and public transit, demonstrated a concerning 95% false positive rate during NYC subway trials
• School districts invested millions in the technology based on claims that it could definitively detect both concealed and openly carried weapons
• Multiple incidents were reported where the scanners failed to detect actual weapons while frequently triggering alarms for innocuous items like water bottles

Regulatory response: The FTC’s settlement with Evolv imposes strict limitations on the company’s marketing practices and provides remedies for affected institutions.

• The company is now prohibited from making unsupported claims about its AI-based weapon detection capabilities
• K-12 schools must be notified of options to cancel certain contracts
• Evolv can no longer make false comparisons between their scanners and traditional metal detectors

Legal implications: Evolv’s practices have triggered broader legal consequences beyond the FTC settlement.

• The company faces investigations into potential securities law violations
• A class-action shareholder lawsuit has been filed
• FTC Chair Lina Khan specifically called out the company for “falsely hyping” its systems to educational institutions

Market impact and public safety concerns: The widespread deployment of Evolv’s technology across sensitive venues raises significant public safety implications.

• The scanners have been installed at airports, subways, theme parks, and film festivals as an alternative to conventional security measures
• NYC subway testing revealed the systems did not demonstrate improved accuracy over time
• The case highlights the risks of overselling AI capabilities as solutions to complex public safety challenges

Looking ahead: This enforcement action signals increasing regulatory scrutiny of AI marketing claims, particularly for technologies deployed in sensitive environments like schools where safety is paramount and the consequences of failure are severe.