CBP facial comparison at Dulles flags a French-passport impostor

CBP's new facial comparison system at Dulles reported a mismatch for a man presenting a French passport on 22 August 2018; a Congolese ID was found in his shoe and he was returned to Brazil.

21ⓘ

AI system

System

CBP facial comparison biometric system (biometric entry/exit, Traveler Verification Service)

Model

Not assessed

Release

Not assessed

Application developer

Not named in the sources

Role

Not assessed

Configuration

Not assessed

Inputs and outputs

Not assessed

Autonomy

Not assessed

Workflow

Live photo at primary inspection compared against passport/visa photos retrieved from government holdings via airline manifests. Officer referral to secondary inspection on a reported mismatch. Traveler using a genuine document issued to someone else.

Human oversight

Not assessed

Relation to AI

The facial comparison match (less than two seconds per CBP) produced the mismatch that triggered secondary inspection. The search and removal were human actions. The documentation is CBP's own release (deployer, not developer) plus NBC with a named CBP official. The algorithm vendor and the match threshold are not in the sources. CBP's “99% match rate” claim is unaudited here.

AI contribution

The facial comparison match (less than two seconds per CBP) produced the mismatch that triggered secondary inspection. The search and removal were human actions. The documentation is CBP's own release (deployer, not developer) plus NBC with a named CBP official. The algorithm vendor and the match threshold are not in the sources. CBP's “99% match rate” claim is unaudited here.

Benefits

Outcome identifiers

Not assessed

Benefit

by September 2021, more than 900 impostors nationally

Category

Reduced Violence

Status

Not assessed

Beneficiary groups

Society. Organizations.

Beneficiaries

Not assessed

Reach

Not assessed

Distribution and access

Not assessed

Duration

Not assessed

Expected?

Yes

Costs and harms

Not assessed

Measurement

Studies

Not assessed

Study design

Not assessed

Measurement

Not assessed

Setting and sample

Not assessed

Measures

by September 2021, more than 900 impostors nationally

Comparison

Not assessed

Results

by September 2021, more than 900 impostors nationally

Additional value from AI

Not assessed

Limitations

Not assessed

Denominators

Not assessed

Materials for replication

Not assessed

Money

Kind of value

Not assessed

Amount

Not assessed

Currency and prices

Not assessed

Period and unit

Not assessed

Realization

Not assessed

Recipients and payers

Not assessed

Gross or net

Not assessed

Included costs

Not assessed

Comparison

Not assessed

Method

Not assessed

Return and payback

Not assessed

People and places

Participants

Developer: Not named in the sources. Operator: U.S. Customs and Border Protection, Office of Field Operations, Washington Dulles International Airport.

Jurisdictions

Benefit: Sterling, United States. Operator: United States.

Ownership and control

Not assessed

Dependencies

Not assessed

Cross-border connections

Not assessed

Geopolitical implications

Not assessed

Follow-up

Conditions for replication

A reference photo of the document's true bearer must exist in DHS holdings. Officers must act on mismatches rather than override them. Photo quality and demographic performance of the matcher must be adequate.

Replication attempts

3 follow-up entries. Benefit recurred: 3.

Later findings

24 September 2021. By September 2021 CBP reported that facial biometrics had stopped more than 900 impostors nationwide since September 2018 and that close to 100 million travelers had been processed.

Availability

Not assessed

Institutional decision

Not assessed

Grounds

Not assessed

Action and result

Not assessed

Success criterion

Not assessed

Evidence

Source numbers

AIBD-0223-source-0001–0003

Citations

Numbered references below.

Language

Not assessed

Source relationships

Operator, Independent

Original evidence

Not assessed

Access and versions

Not assessed

Supported findings

Not assessed

Evidence strength

Not assessed

Missing evidence

Not assessed

Grades

Assessed benefits

Society. Organizations.

Depth

Society: 2. Organizations: 1.

Reasons

Society: An attempt to enter the United States on another person's genuine passport was stopped and the traveler removed. Sources document no further threat beyond the document fraud. Organizations: CBP gained a verified first detection for a system then three days into use at Dulles. Subsequent releases report repeated catches.

Observed or intended

Not assessed

Rubric

Not assessed

Assessor and date

Not assessed

Alternative assessments

Not assessed

Fill

Independently reported

Entry details

Kind

Not assessed

Field

Border security. Government.

Purpose

Verify arriving international travelers' identities by comparing a live photo with passport or visa photos in government holdings, to detect impostors and expedite entry.

Dates

Date basis unassessed: 2018 (precision unassessed). Updated: 21 October 2026.

Stage

Not assessed

Version

SHA-256: db88b80b0b25…

References

Other identifiers

Not assessed

Category mappings

Not assessed

Brief

Not assessed

Revisions

5 revisions. Latest: 21 October 2026.

Sources

  1. News report Independent author

    Tom Costello and Ethan Sacks, NBC News, 23 Aug 2018

  2. Press release Deployer

    U.S. Customs and Border Protection, local media release, 23 Aug 2018

  3. Press release Deployer

    U.S. Customs and Border Protection, local media release, 24 Sep 2021