AI Retail Surveillance & False Shoplifting Accusations: Legal Guide

When Store AI Accuses Innocent Shoppers

#

Retailers are deploying artificial intelligence surveillance systems at unprecedented scale—and innocent customers are paying the price. From facial recognition cameras at self-checkout kiosks to AI-powered “theft prediction” systems, stores are using technology that falsely accuses shoppers of crimes they didn’t commit. The consequences include public humiliation, wrongful detention, false arrests, and permanent damage to reputations.

The Federal Trade Commission banned Rite Aid from using facial recognition for five years after finding the system generated thousands of false positives, disproportionately targeting Black and Asian customers. Home Depot faces class action lawsuits over allegedly collecting facial data at self-checkout without consent. Target is fighting BIPA claims that it captures shoppers’ biometric information through surveillance systems. And across the country, innocent people are being detained, searched, and arrested based on AI matches that turn out to be wrong.

How Retail AI Surveillance Works

#

Facial Recognition Systems

#

Modern retail surveillance goes far beyond security cameras. AI-powered systems analyze video in real time to:

Identify “Known Shoplifters”:

• Compare customer faces against internal databases
• Cross-reference with law enforcement watchlists
• Track individuals across multiple store visits
• Flag customers based on prior accusations (proven or not)

Predict “Suspicious” Behavior:

• Analyze body language and movement patterns
• Detect “concealment gestures”
• Monitor shopping cart loading patterns
• Flag customers who pick up multiple items

The Technology:

• “Computer vision” AI analyzes geometric facial points
• Creates unique “faceprints” stored in databases
• Compares live video against stored templates
• Generates alerts for store employees to act on

Self-Checkout Surveillance

#

Self-checkout kiosks have become a surveillance flashpoint:

AI Monitoring Systems:

• Cameras above kiosks track item scanning
• Weight sensors verify items placed in bags
• AI compares scanned items to video of cart contents
• “Missed scan detection” flags potential theft

Common Systems:

• Everseen: Used by Walmart at thousands of stores; employees call it “NeverSeen” due to frequent errors
• Computer vision platforms: Match items seen to items scanned
• Receipt verification AI: Compares purchase to exit surveillance

The Problem: These systems produce massive false positive rates. Items that don’t scan properly, produce weighed incorrectly, or customers who simply move too fast can all trigger alerts—leading employees to accuse, detain, or call police on innocent shoppers.

The Human Cost

#

When AI surveillance fails, employees typically:

• Follow and surveil flagged customers
• Publicly confront and accuse them
• Detain them in back rooms
• Search their belongings
• Call police to remove them
• Issue trespass bans
• Seek prosecution

All based on an algorithm that may be wrong 90%+ of the time.

The Rite Aid FTC Action

#

What Happened

#

In December 2023, the Federal Trade Commission announced a landmark settlement with Rite Aid over its facial recognition surveillance program.

FTC Findings:

How Employees Responded to AI Alerts:

• Followed consumers around stores
• Searched customers and their belongings
• Ordered people to leave
• Called police to confront or remove customers
• Publicly accused people of shoplifting
• Accused people in front of friends and family

Disproportionate Impact: The FTC specifically found Rite Aid’s technology was “more likely to generate false positives in stores located in predominantly Black and Asian neighborhoods than in predominantly white communities.”

Settlement Terms

#

Five-Year Facial Recognition Ban: Rite Aid is prohibited from using facial recognition technology for security or surveillance purposes for five years.

Data Destruction Requirements:

• Must destroy all photos and videos used in facial recognition systems
• Must instruct third parties to destroy all models or algorithms developed using wrongly collected images

Information Security Program:

• Must implement robust information security
• Oversight required from top executives
• Third-party service provider oversight mandated

Home Depot BIPA Lawsuit

#

The Allegations

#

In August 2025, a Chicago resident filed a class action lawsuit against Home Depot in the U.S. District Court for the Northern District of Illinois, alleging the company secretly uses facial recognition at self-checkout kiosks without customer consent.

Key Claims:

• Cameras above self-checkout kiosks display green boxes around customer faces
• Facial recognition scans geometric points to create stored “faceprints”
• No signage notifies customers of biometric data collection
• No consent obtained before collecting facial data
• Technology deployed at all 76 Illinois Home Depot locations

How It Was Discovered: Lead plaintiff Benjamin Jankowski noticed a camera above the kiosk displaying a green box around his face while using self-checkout—suggesting active facial scanning was occurring.

Home Depot’s Technology:

• August 2023: Announced use of “computer vision” AI in stores
• May 2024: Expanded technology to combat self-checkout theft
• Uses AI to derive information from digital images and video

Potential Damages

#

Under Illinois’ Biometric Information Privacy Act:

With 76 Illinois locations and potentially millions of customer visits, damages could reach into the hundreds of millions if plaintiffs prevail.

Target BIPA Litigation

#

Arnold v. Target Corporation

#

Court: U.S. District Court, Northern District of Illinois Case No.: 1:24-cv-04452 Status: Proceeding (motion to dismiss denied November 2024)

Allegations:

• Target uses in-store security cameras to capture facial geometry
• Surveillance systems create and store biometric data
• No written policy regarding biometric data collection
• No written consent obtained from customers
• Biometric data shared with third parties

Target’s Surveillance Network: According to the complaint, Target operates:

• 14 investigation centers nationwide
• Two forensic labs to enhance video and analyze fingerprints
• Advanced electronic surveillance capturing faces at every entry/exit

Court Ruling (November 2024): Judge denied Target’s motion to dismiss, finding plaintiffs presented plausible claims that Target uses facial recognition-enabled surveillance in violation of BIPA.

Self-Checkout False Accusations

#

The Scale of the Problem

#

Self-checkout theft has become a retail crisis—but so have false accusations against innocent customers.

Theft Statistics:

• Self-checkout lanes experience 3.5-4% shrink rates
• Staffed registers: approximately 0.21% shrink
• That’s up to 65% higher theft risk at self-checkout

False Accusation Drivers:

• Machines that don’t register scans properly
• Weight sensor errors
• System freezes during transactions
• Card reader failures
• AI misinterpreting normal behavior

Landmark Cases

#

Lesleigh Nurse v. Walmart ($2.1 Million Verdict): A Mobile County, Alabama jury unanimously awarded $2.1 million to a woman falsely accused of shoplifting at a Walmart self-checkout in 2016. Nurse was at self-checkout with her husband and three kids when a malfunctioning scanner caused problems. Despite getting help from a Walmart associate, she was accused of theft.

Expert testimony revealed: In a two-year period, Walmart charged approximately 1.4 million people nationwide with criminal theft of property and collected more than $300 million through civil demand letters.

Harvey Murphy Jr. v. Macy’s/Sunglass Hut ($10M Demanded): A Texas man sued after facial recognition technology falsely identified him as an armed robber. Murphy was held in jail for nearly two weeks before prosecutors verified he wasn’t even in the state during the robbery. During detention, Murphy was allegedly attacked, leaving permanent injuries.

Legal Claims for False Accusations

#

Against the Retailer

#

False Imprisonment/False Arrest:

• Unlawful restraint of liberty
• No need to prove handcuffs or locked rooms
• Even temporary detention can constitute false imprisonment
• Key question: Was detention reasonable and based on probable cause?

Defamation:

• Public accusation of shoplifting damages reputation
• Accusations in front of other customers, friends, or family
• Store communications to law enforcement
• Statements to civil recovery companies

Negligence:

• Failure to properly train loss prevention staff
• Over-reliance on flawed AI systems
• Failure to verify AI alerts before acting
• Failure to implement reasonable safeguards

Intentional Infliction of Emotional Distress:

• Extreme and outrageous conduct
• Public humiliation and accusations
• Particularly strong for pregnant women, elderly, disabled
• Detention in degrading conditions

BIPA Violations (Illinois):

• Collecting biometric data without consent
• Failing to provide written policy
• Sharing biometric data with third parties
• $1,000-$5,000 per violation

Against AI Vendors

#

Product Liability:

• AI system defectively designed
• Known high false positive rates
• Failure to warn about racial disparities
• Failure to provide adequate safeguards

Negligence:

• Failure to test for accuracy
• Failure to validate against racial bias
• Inadequate training data
• Failure to disclose limitations

BIPA and Biometric Privacy

#

Illinois BIPA

#

The Illinois Biometric Information Privacy Act provides the strongest biometric privacy protections in the nation:

Requirements:

• Written policy on retention and destruction
• Informed consent before collection
• No sale, lease, or trading of biometric data
• Reasonable security measures

Private Right of Action:

• $1,000 per negligent violation
• $5,000 per intentional/reckless violation
• Attorney’s fees to prevailing plaintiffs
• Class actions permitted

2024 Amendments:

• Damages now calculated “per person” not “per scan”
• Reduces potential for “ruinous” damage awards
• Still provides significant penalties

Other State Laws

#

Major BIPA Settlements

#

What to Do If Falsely Accused

#

Immediate Steps

#

During the Incident:

1. Stay calm — Don’t argue, resist, or escalate
2. Ask if you’re free to leave — Clarifies whether you’re detained
3. Don’t sign anything — Especially “civil recovery” acknowledgments
4. Don’t make statements — You have the right to remain silent
5. Request to call an attorney — Before answering questions
6. Note names and badge numbers — Of employees involved
7. Look for witnesses — Other customers who saw what happened

Immediately After:

8. Get incident report — Request a copy from the store
9. Preserve your receipt — Proof of legitimate purchase
10. Photograph everything — Your receipt, items, the store
11. Write down details — While fresh in memory
12. Request surveillance video — Send written preservation demand
13. Don’t return to the store — Avoid potential confrontations

Evidence Preservation

#

Send a written preservation letter to the retailer demanding they preserve:

If Police Were Involved

#

If Arrested:

• Exercise right to remain silent
• Request an attorney
• Don’t consent to searches
• Document everything

After Release:

• Get copy of police report
• Request body camera footage
• Note officer names and badge numbers
• File complaint if rights violated

If Charges Filed:

• Hire criminal defense attorney
• Gather alibi evidence
• Obtain surveillance video
• Document all damages for civil claim

Damages and Recovery

#

What You Can Recover

#

Compensatory Damages:

• Lost wages from detention, arrest, court appearances
• Legal fees for criminal defense
• Medical expenses (physical and mental health)
• Damage to reputation
• Loss of employment opportunities

Emotional Distress:

• Humiliation from public accusation
• Anxiety and depression
• PTSD from wrongful arrest
• Impact on family relationships

Special Damages:

• Costs of restoring reputation
• Employment background check issues
• Immigration consequences
• Professional license impacts

Punitive Damages:

• Available for intentional or reckless conduct
• Stronger where retailer knew AI was flawed
• Pattern of false accusations supports punitive awards

Settlement Benchmarks

#

Frequently Asked Questions

#

Related Resources

#

• AI Facial Recognition Wrongful Arrests — Police facial recognition errors
• AI Tenant Screening Discrimination — AI bias in housing
• AI Insurance Claim Denials — Algorithmic claim denials
• Understanding Liability — Legal frameworks for AI harm
• Filing a Claim — Step-by-step claims process