← Back to Blog

US-VISIT: the border-exit system Congress ordered in 1996 and still has not fully delivered

· 11 min read US-VISIT: the border-exit system Congress ordered in 1996 and still has not fully delivered

When a foreign traveler lands at a U.S. airport and presses their fingers onto a scanner, they are using a system that works. That is the entry half of US-VISIT, and it has been operating at nearly every port of entry since 2004. What that same traveler does not encounter on the way out, at least not at a land crossing and not in the comprehensive form Congress ordered, is the matching exit check. The exit half is the part America paid for repeatedly, mandated by law again and again, and never finished as designed.

This is not a story of a program that failed outright. It is a story of a half-built machine. One half runs. The other half, the one Congress first ordered in 1996, has been under construction, redefined, reorganized, and partially rebuilt for more than two decades, and at land borders, where most travel happens, it still largely does not exist. The government's own auditors documented both halves in detail, and the honest version keeps them distinct.

What US-VISIT was, and what it was supposed to do

The United States Visitor and Immigrant Status Indicator Technology program, known as US-VISIT, was established by the Department of Homeland Security in 2003. Its first biometric phase went live through an interim final rule published in the Federal Register on January 5, 2004, requiring covered foreign travelers to give two fingerprints and a photograph on arrival and departure at air and sea ports of entry. A second rule in August 2004 extended coverage to Visa Waiver travelers and designated land ports.

The design was symmetric on paper: capture biometrics coming in, capture biometrics going out, and reconcile the two so the government would know who had actually left and who had overstayed a visa. The entry side was delivered. The exit side, especially the reconciliation Congress wanted, is where the record turns difficult.

It matters that the exit requirement did not begin with US-VISIT, and did not begin with the attacks of September 11, 2001. The mandate for an automated entry-exit system originated in the Illegal Immigration Reform and Immigrant Responsibility Act of 1996, at Section 110, which called for a system recording the arrivals and departures of noncitizens, originally due in 1998. That 1996 requirement was biographic and automated, not yet biometric. The biometric requirement was layered on later, after 9/11: the USA PATRIOT Act of 2001 (Section 414) directed the use of biometric technology, and the Enhanced Border Security and Visa Entry Reform Act of 2002 (Section 302, codified at 8 U.S.C. 1365b) required an integrated biometric entry-exit system. Between those, the Data Management Improvement Act of 2000 reset deadlines. In short, the exit obligation was written in 1996, made biometric around 2001 and 2002, and reaffirmed by Congress across administrations. US-VISIT, launched in 2004, was the vehicle built to satisfy a mandate that already predated it by eight years.

What the record establishes

The most direct account comes from the Government Accountability Office. In testimony delivered in June 2007 (GAO-07-1044T), GAO stated that DHS had, in its words, "invested about $1.3 billion over 4 years and delivered basically one-half of US-VISIT." The delivered half was biometric entry, operating at what GAO described as almost 300 air, sea, and land ports of entry, screening arriving travelers against terrorist and criminal watch lists. On the other half, GAO was blunt: "DHS still does not have an operational exit capability."

That $1.3 billion figure was the amount DHS had obligated as of January 31, 2007, according to program officials. It is a snapshot of that era, not a lifetime total. GAO's companion work put appropriations at about $1.7 billion through fiscal year 2007. Of the exit money specifically, GAO reported that DHS allocated roughly $250 million to exit-related efforts between 2003 and 2006, described in the testimony as about one-quarter of a billion dollars, and that this spending produced no operational biometric exit.

Why not? GAO recorded the reason in the words of US-VISIT's own officials. Mirroring the entry inspection at land ports of entry was judged "an infeasible alternative for numerous reasons, including but not limited to, the additional staffing demands, new infrastructure requirements, and potential trade and commerce impacts." An interim technology tested at land borders was not biometric and therefore did not meet the statutory requirement. This is the structural fact underneath the whole story: the United States built inspection infrastructure to check people coming in, but it never built physical exit-control infrastructure to check people going out. At an airport, an airline controls the gate and there is a natural chokepoint. At a land border, there is often nothing on the departure side at all.

The timeline

  • 1996: IIRIRA Section 110 requires an automated entry-exit system recording noncitizen arrivals and departures, originally due 1998.
  • 2000: The Data Management Improvement Act resets deadlines for air, sea, and the busiest land ports.
  • 2001: The USA PATRIOT Act, Section 414, directs use of biometric technology.
  • 2002: The Enhanced Border Security and Visa Entry Reform Act, Section 302, requires an integrated biometric entry-exit system, codified at 8 U.S.C. 1365b.
  • 2003: DHS establishes US-VISIT.
  • January 5, 2004: Interim final rule launches biometric entry and exit collection at air and sea ports; an August 2004 rule extends coverage.
  • June 2007: GAO reports about $1.3 billion obligated, "basically one-half" of US-VISIT delivered, and no operational exit capability. Roughly $1.7 billion appropriated through FY2007, about $250 million of it on exit efforts since 2003.
  • March 2013: US-VISIT is reorganized into the Office of Biometric Identity Management (OBIM) under the FY2013 Homeland Security Appropriations Act, appropriated about $232 million. Overstay analysis moves to Immigration and Customs Enforcement; biometric exit development moves to Customs and Border Protection. OBIM is divested of the exit and overstay-rate missions.
  • December 2015: The Consolidated Appropriations Act, 2016 (Public Law 114-113, Section 411) authorizes H-1B and L-1 visa fee increases to provide up to $1 billion for a biometric exit system beginning in FY2017.
  • May 2020: CBP has deployed facial-recognition air-exit technology at 27 airports (GAO-20-568); sea and land ports remain in early-stage assessment, and comprehensive biometric land exit is not operationally deployed.

The money, with its dates attached

The dollar figures in this story are easy to blur together, and blurring them produces a number that never existed. Kept distinct, they read like this.

  • About $1.3 billion: obligated by DHS for US-VISIT as of January 31, 2007, per GAO-07-1044T. This is the figure paired with GAO's "basically one-half" finding.
  • About $1.7 billion: appropriated for US-VISIT through fiscal year 2007, per GAO's broader review (GAO-07-248). A different measure of the same era, not additive to the $1.3 billion.
  • About $250 million: allocated to exit-related efforts between 2003 and 2006, with no operational biometric exit fielded.
  • About $232 million: appropriated in the FY2013 reorganization that created OBIM.
  • Up to $1 billion: the cap authorized in December 2015 for biometric exit, funded not by a direct appropriation but by increased H-1B and L-1 visa fees collected by U.S. Citizenship and Immigration Services through FY2027. Two cautions belong on this one. First, the same Section 411 fee also funds 9/11 first-responder healthcare, so the fee is not all exit money. Second, collection has run below target; the up-to-$1 billion is a ceiling, not a bank balance, and later spending plans projected collecting well under the full amount by FY2027.

None of these is a program-lifetime total, and they should not be summed into one. The through-FY2007 figures are early snapshots; OBIM and CBP spending on biometric identity and exit work in the years since is separate and larger.

Where the exit mission actually went

The tense here matters, because it is easy to overstate the failure. It is not accurate to say biometric exit was simply never built. What was not delivered was the comprehensive, mandated system, and above all land exit.

On the air side, the mission advanced later, where it was tractable. As of May 2020, GAO reported (GAO-20-568) that CBP, working with airline partners, had deployed facial-recognition technology to biometrically confirm departing travelers at 27 airports. That is a real capability, and it grew after 2020. It is also not the fingerprint-based exit originally envisioned; it uses face matching, and GAO flagged privacy and system-performance shortcomings that are worth noting on their own terms and should not be folded into the delivery-and-cost critique.

On the sea and land sides, the picture in 2020 was thinner. GAO described CBP as in the early stages of assessing facial-recognition technology at sea and land ports, not operating it. Comprehensive biometric land exit, the piece that would cover the borders where most departures and most overstay risk concentrate, remained undeployed. That is the enduring gap in a mandate first written in 1996.

One clarification on more recent auditing. GAO returned to the broader biometric enterprise in 2023 (GAO-23-105959) and found significant shortcomings in program management and privacy, including a 33-month schedule slip and a roughly $354 million cost rebaseline. Those specific findings belong to HART, the Homeland Advanced Recognition Technology database that is replacing the older IDENT identity system, not to the exit program itself. The 2023 report shows the biometric enterprise is still troubled; it is not, on its own, a fresh indictment of the exit build, and it should not be cited as one.

The honest failure critique and the honest mission defense

The failure critique. Congress ordered an automated entry-exit system in 1996, made it biometric after 9/11, and reaffirmed the requirement repeatedly. DHS stood up US-VISIT and, by GAO's June 2007 account, had obligated about $1.3 billion and appropriated roughly $1.7 billion through FY2007, yet delivered "basically one-half" of the program. About $250 million went specifically to exit efforts from 2003 to 2006 with nothing operational to show for it. The program was reorganized into OBIM in 2013 and stripped of the exit mission, which passed to CBP; Congress then dedicated up to $1 billion in visa-fee funding in 2015 to try again, and even that has under-collected. More than two decades and multiple funding streams after the original mandate, the comprehensive biometric exit system Congress ordered, especially at land borders, has not been delivered as designed. By the standard of what was legislated and paid for, that is a long-running shortfall.

The mission defense. The delivered half is genuine and valuable. Biometric entry runs at nearly 300 ports of entry, capturing fingerprints and photographs from millions of arriving travelers and checking them against watch lists, a capability that simply did not exist before 2004. The unbuilt half reflects physical and engineering reality more than incompetence. The United States has never operated universal exit controls and built no exit infrastructure at land borders; replicating entry inspection there would require new facilities, more staff, and would risk throttling cross-border trade and commerce, which is why DHS's own officials called land biometric exit infeasible as configured. And the mission was not abandoned. Where it was achievable, CBP moved: facial-recognition air exit reached 27 airports by 2020 and expanded afterward, with sea and land under assessment. That is measured progress on the tractable portions, alongside a hard problem that money alone was never going to solve.

Both verdicts are true at once. A working entry system that screens millions is a real accomplishment. A mandated exit system that remained comprehensively undelivered for more than twenty years, especially at land borders, is a real failure of delivery. The instructive part is why the gap persisted: not a single scandal, but a mandate that outran the physical infrastructure and the political appetite to build it, restated by Congress far more often than it was funded to completion.

Fact-check notes and sources

  • US-VISIT established 2003; first biometric phase via interim final rule published January 5, 2004, covering fingerprints and photograph on arrival and departure at air and sea ports (a second rule August 31, 2004 added Visa Waiver travelers and land ports): Federal Register, DHS, Jan 5, 2004.
  • DHS "invested about $1.3 billion over 4 years and delivered basically one-half of US-VISIT," and "DHS still does not have an operational exit capability"; entry operating at almost 300 ports of entry; land biometric exit called "an infeasible alternative" for staffing, infrastructure, and trade/commerce reasons; about $250 million on exit efforts 2003 to 2006. The $1.3 billion is obligated as of January 31, 2007: GAO-07-1044T, June 28, 2007.
  • About $1.7 billion appropriated for US-VISIT through fiscal year 2007: GAO-07-248, 2007.
  • Exit-tracking mandate originated in IIRIRA 1996 Section 110 (automated entry-exit system, originally due 1998), predating US-VISIT; deadlines repeatedly extended: CRS IF11634, Biometric Entry-Exit System: Legislative History and Status.
  • Mandate amended by the Data Management Improvement Act of 2000, the USA PATRIOT Act of 2001 (Section 414, biometric technology), and the Enhanced Border Security and Visa Entry Reform Act of 2002 (Section 302), codified at 8 U.S.C. 1365b.
  • 2013 reorganization of US-VISIT into OBIM (about $232 million appropriated), with overstay identification moving to ICE and biometric exit development to CBP: DHS Office of Biometric Identity Management overview (secondary summary of the FY2013 Homeland Security Appropriations Act; treat the appropriations act itself and DHS.gov/obim as the primary basis).
  • Consolidated Appropriations Act, 2016 (Public Law 114-113, Section 411, December 2015) authorized H-1B and L-1 fee increases to provide up to $1 billion for biometric exit beginning FY2017; the same fee also funds 9/11 first-responder healthcare, and collection has lagged: CRS R47541, Immigration: The U.S. Entry-Exit System.
  • Facial-recognition air exit deployed at 27 airports as of May 2020; sea and land ports in early-stage assessment; comprehensive biometric land exit not operationally deployed; privacy and performance shortcomings flagged: GAO-20-568, September 2020.
  • Continued biometric-program shortcomings in 2023, including a 33-month schedule slip and roughly $354 million cost rebaseline, pertain to the HART identity database replacing IDENT, not to the exit program specifically: GAO-23-105959, September 2023.

Related reading

This post is informational and journalistic, not legal or financial advice. It describes public programs and documented events; mentions of third parties are nominative fair use and no affiliation is implied.

← Back to Blog

Accessibility Options

Text Size
High Contrast
Reduce Motion
Reading Guide
Link Highlighting
Accessibility Statement

J.A. Watte is committed to ensuring digital accessibility for people with disabilities. This site conforms to WCAG 2.1 and 2.2 Level AA guidelines.

Measures Taken

  • Semantic HTML with proper heading hierarchy
  • ARIA labels and roles for interactive components
  • Color contrast ratios meeting WCAG AA (4.5:1)
  • Full keyboard navigation support
  • Skip navigation link
  • Visible focus indicators (3:1 contrast)
  • 44px minimum touch/click targets
  • Dark/light theme with system preference detection
  • Responsive design for all devices
  • Reduced motion support (CSS + toggle)
  • Text size customization (14px–20px)
  • Print stylesheet

Feedback

Contact: jwatte.com/contact

Full Accessibility StatementPrivacy Policy

Last updated: April 2026