The Claim

Launched the ICE “Worst of the Worst” database, publicly identifying criminal illegal aliens.

The Claim, Unpacked

What is literally being asserted?

That the administration launched a database — publicly accessible — that identifies individuals described as “criminal illegal aliens.” The factual core is narrow: a database was created and made public.

What is being implied but not asserted?

That the database contains the “worst of the worst” — people who pose the most serious threats to public safety. That publicly identifying these individuals serves a legitimate public safety purpose analogous to sex offender registries. That the people listed are all convicted criminals. That the database is accurate. That publicly listing noncitizens by name, photo, and alleged crime is a straightforward transparency measure with no significant civil liberties implications.

What is conspicuously absent?

Any mention that independent analysis found 56% of the people listed on the database had not been charged with or convicted of a violent crime. Any disclosure that the database includes people charged but not convicted — listing them publicly alongside convicted murderers and sex offenders despite the presumption of innocence. Any acknowledgment that DHS itself admitted the database was “rife with errors,” with charges for hundreds of people described incorrectly, including at least one case where the White House publicly accused a man of a child sex crime he did not commit. Any disclosure that the 10,000 initial entries represent less than 5% of total ICE arrests — meaning 95% of the people ICE arrested did not make the “worst of the worst” cut, and even within that 5%, more than half had no violent charges. Any mention that the Washington Post’s investigation of thousands of internal ICE messages revealed the “Worst of the Worst” label was part of a coordinated viral social media strategy designed by the White House and executed by DHS staff including a lifestyle influencer, a wedding videographer, and a cable-TV actor. Any discussion of whether publicly listing people — including those only charged, not convicted — on a government website constitutes a form of extrajudicial punishment without due process. Any acknowledgment that ProPublica documented over 170 U.S. citizens wrongfully detained by ICE, raising fundamental questions about the accuracy of the agency’s identification of who is and is not a “criminal illegal alien.”

Evidence Assessment

Established Facts

The WOW.DHS.GOV database was launched on December 8, 2025. DHS announced the launch of a new webpage called “Worst of the Worst” at wow.dhs.gov, described as a searchable website that “lets you find criminal illegal aliens removed from your community.” The site launched with approximately 10,000 entries and has been expanded regularly: to 15,000 by December 18, 2025; 20,000 by January 22, 2026; 25,000 by February 5, 2026; and over 30,000 by February 20, 2026. The database allows users to search by state and view individuals’ names, countries of origin, crimes they are accused of or convicted of, and the communities they were arrested in. DHS Assistant Secretary for Public Affairs Tricia McLaughlin stated the purpose was “so every American can see for themselves the criminal illegal aliens that we are arresting, what crimes they committed, and which communities we removed them from.” [^037-a1]

The database includes people who are merely charged, not convicted. DHS’s own language describes the listed individuals as “charged or convicted of a crime.” This means the database publicly identifies people who have not been found guilty of any offense, alongside people with serious criminal convictions. Immigration law scholars have documented that the presumption of innocence is “completely absent from quasi-criminal removal proceedings” — a not guilty verdict is treated as “virtually meaningless” in immigration adjudications. The WOW database extends this absence of presumption into the public sphere: people who may ultimately be acquitted are nevertheless publicly listed on a government website as “criminal illegal aliens.” [^037-a2]

DHS admitted the database contained widespread errors. CNN’s February 2026 investigation found that charges against hundreds of people listed on the website were described incorrectly. DHS acknowledged the errors, attributing them to a “glitch” affecting approximately 5% of entries — which, at 25,000 entries, means approximately 1,250 people were publicly listed with incorrect criminal charges on a government website. In a separate incident reported by NOTUS, the White House conceded it posted a picture of a man who was erroneously described as having been convicted of a sex crime involving a child. DHS claimed the glitch was “resolved” by February 18, 2026, but refused to specify what the glitch was. [^037-a3]

56% of the people listed have not been charged with or convicted of a violent crime. David Bier of the Cato Institute independently analyzed the database and found “a majority (56 percent) of the list has not been charged or convicted of a violent crime.” Nearly one-quarter (25%) faced only vice, immigration, or non-DUI traffic charges. The database includes people arrested for drug possession, writing bad checks ($80 total from a decade prior), immigration violations, and other nonviolent infractions. Reason documented specific cases: a Purple Heart veteran (Sae Joon Park) listed for a 2007 drug possession conviction, and a Navy veteran’s wife — an Irish immigrant of 48 years — facing deportation for two bad checks from a decade ago. [^037-a4]

The “worst of the worst” represent less than 5% of total ICE arrests. At launch, the database contained approximately 10,000 entries against roughly 281,000 total ICE arrests since January 20, 2025 — meaning DHS itself classified less than 4% of its arrests as “worst of the worst.” Aaron Reichlin-Melnick of the American Immigration Council observed: “DHS is implicitly admitting that less than 5% of the people it arrests are people they believe are ‘the worst of the worst.’” Even this small subset includes the 56% who have no violent charges, meaning the truly violent offenders on the list represent approximately 2% of total ICE arrests. [^037-a5]

Strong Inferences

The database is a messaging tool, not a public safety tool. The Washington Post reviewed thousands of internal ICE messages revealing that the “Worst of the Worst” branding was part of a coordinated viral social media strategy. Internal messages showed ICE officials stating “we need to be highlighting worst of worst LA arrests across relevant accounts/channels.” ICE posted 38 tweets over 11 hours in one session. Video producers accompanied officers on raids and were expected to alert supervisors about “particularly cinematic” scenes. DHS hired a MAGA women’s lifestyle influencer, an L.A.-based wedding videographer, and a Canadian-born cable-TV actor as public affairs staff to produce content. The database — and its “Worst of the Worst” label — was designed to serve a narrative purpose: making enforcement operations look like they exclusively target dangerous criminals, when the data shows the opposite. [^037-a6]

The database functions as extrajudicial public shaming without due process protections. Sex offender registries — the closest analogue — have faced extensive constitutional challenges. Courts have found sex offender registries unconstitutional under the cruel and unusual punishment clause (Colorado, 2017) and under ex post facto provisions (Pennsylvania Supreme Court, 2017). Sex offender registries at least require convictions. The WOW database goes further: it publicly lists people who are merely charged, broadcasting allegations “without the kind of context that typically emerges later in court proceedings.” Listed individuals face permanent reputational harm in workplaces, schools, and communities. For people whose charges are later dropped or who are acquitted, the damage from public listing on a government website labeled “Worst of the Worst” cannot be undone. The database provides no mechanism for listed individuals to contest their inclusion or correct errors. [^037-a7]

ICE’s documented pattern of misidentification makes the database’s accuracy inherently suspect. ProPublica documented over 170 U.S. citizens wrongfully detained by ICE, with agents regularly rejecting legitimate proof of citizenship. Congressional investigators found ICE officers dragged people from cars, detained them for days, fabricated claims of assault, and denied medical care. DHS spokesperson stated “We don’t arrest US citizens for immigration enforcement” — a claim contradicted by over 170 documented cases. If ICE cannot reliably determine who is and is not a citizen, its ability to accurately classify and publicly list individuals as “criminal illegal aliens” on a searchable government website is fundamentally compromised. [^037-a8]

The database’s relationship to VOICE (Item #27) creates a mutually reinforcing narrative infrastructure. VOICE exists to institutionalize the premise that immigrant crime is a distinct and significant category. The WOW database provides the visual evidence to support that premise — mugshots, names, crime descriptions — even though the data shows most entries involve nonviolent offenses and many involve only charges, not convictions. Together, VOICE and WOW create a feedback loop: VOICE presupposes that immigrant crime is a distinctive threat; WOW provides the curated imagery to make that threat feel visceral and immediate; both exist primarily as narrative infrastructure rather than functional public safety tools. [^037-a9]

Informed Speculation

The database’s design reveals its true function. A legitimate public safety database would include only individuals convicted of serious crimes who pose an ongoing danger — similar to how sex offender registries (despite their own problems) are at least nominally tied to ongoing risk. The WOW database includes people who have been removed from the country, people with decades-old minor offenses, people who are merely charged, and people whose charges were incorrectly listed by DHS itself. None of these inclusions serve a public safety purpose. They serve a narrative purpose: generating a large, searchable repository of mugshots and crime descriptions that reinforces the association between immigration and criminality.

The 5% “glitch” rate is worth contextualizing. DHS claimed approximately 5% of entries had incorrect charges. At 25,000 entries, that is roughly 1,250 people publicly and incorrectly labeled on a government website as having committed crimes they did not commit. For an ordinary citizen, being falsely listed on a government “criminal” database would be considered a serious violation. For noncitizens in removal proceedings, the lack of any meaningful redress mechanism means the reputational harm is permanent and compounding — potential employers, landlords, and community members who search WOW.DHS.GOV will find incorrect information that DHS has no obligation to correct proactively.

The database’s growth pattern — adding 5,000 entries at regular intervals, each announced with a taunting press release — suggests the primary metric of success is database size, not database accuracy. The February 20, 2026 announcement (“To Award the Mainstream Media for Raising Awareness”) was issued the day after CNN reported the database was “rife with errors,” suggesting DHS responded to criticism about accuracy by adding more entries.

Structural Analysis

Stated vs. revealed preferences. The administration states the database serves “transparency” and public safety. The revealed preference is viral content production. Internal ICE messages document a coordinated effort to use the “Worst of the Worst” brand as social media content — complete with lifestyle influencers and wedding videographers producing “cinematic” footage. If the purpose were genuinely public safety, the database would include only convicted violent offenders, would exclude people who have already been deported, and would include an accuracy review and error correction mechanism. Instead, it includes people charged with writing $80 in bad checks, grows by 5,000 entries at regular media-cycle intervals, and had a 5% error rate that DHS acknowledged only when pressed by CNN.

The padding lens. Listing the database as a separate “win” (Item #37) alongside “650,000 arrests, detentions, and deportations” (Item #3), “400,000 criminal aliens deported” (Item #4), and VOICE restoration (Item #27) demonstrates how a single enforcement operation generates multiple list entries. The database draws from the same pool of arrests already counted in Items #3 and #4. It serves the same narrative function as VOICE (Item #27). Four separate “wins” describe different presentations of the same underlying enforcement activity.

Cui bono. The database benefits the administration’s political messaging by creating a permanent, searchable visual archive of immigrant criminality. Each entry — with its mugshot, name, and crime description — functions as a micro-narrative reinforcing the association between immigration and danger. The database is designed to be shared on social media, embedded in news stories, and searched by individual communities. The political beneficiary is the administration’s immigration narrative. The harmed parties are: (a) the individuals listed incorrectly or for minor offenses; (b) immigrant communities subjected to heightened suspicion; and (c) the principle that government should not publicly brand people as criminals before conviction.

The presumption of innocence problem. The database includes people who are “charged or convicted.” In the American legal system, charges are accusations — not findings of guilt. Publicly listing a person on a government website labeled “Worst of the Worst” based on charges alone constitutes a government determination of guilt disseminated to the public before any judicial proceeding has occurred. While immigration proceedings are technically civil, not criminal, the consequences of public listing — reputational destruction, employment barriers, community ostracism — are functionally punitive. Immigration law scholars have documented that the presumption of innocence is systematically absent from immigration enforcement; the WOW database extends this absence from courtrooms into the public sphere.

Context the Framing Omits

Less than 14% of ICE arrests in the first year involved people with violent criminal records. CBS News obtained an internal DHS document showing only 13.9% of approximately 393,000 ICE arrests involved individuals charged with or convicted of violent crimes. Less than 2% had homicide or sexual assault charges. Nearly 40% had no criminal record at all. The “Worst of the Worst” label describes a small fraction of a small fraction of the enforcement population — but the database presents this fraction as representative.

73.6% of ICE detainees had no criminal conviction. TRAC (Syracuse University) data as of February 2026 shows that 50,259 out of 68,289 people in ICE detention — 73.6% — had no criminal conviction. The number detained with no criminal record grew from 3,165 in February 2025 to 25,193 in January 2026 — a 2,450% increase. The enforcement expansion has overwhelmingly targeted people with no criminal history, not the “worst of the worst.”

ICE has repeatedly arrested U.S. citizens. ProPublica documented over 170 cases of U.S. citizens detained by ICE, including children. Congressional investigators confirmed patterns of excessive force, fabricated charges, and rejection of legitimate citizenship documents. If ICE cannot reliably identify who is a citizen, the accuracy of its public criminal database — which requires correct identification of both immigration status and criminal history — is fundamentally unreliable.

The database admitted a 5% error rate only when forced. DHS acknowledged errors in the database only after CNN’s investigation. The “glitch” affected approximately 1,250 people at the 25,000-entry mark — people publicly and incorrectly labeled as having committed crimes they did not commit. DHS’s response was to add more entries, not to audit existing ones.

The database has no error correction or removal mechanism for listed individuals. Unlike sex offender registries, which include (limited) mechanisms for review and removal, the WOW database provides no process for listed individuals to challenge their inclusion, correct errors, or request removal after charges are dropped or cases are dismissed.

Verdict

Factual core: True. The administration did launch the WOW.DHS.GOV database on December 8, 2025. It is publicly accessible, searchable, and contains over 30,000 entries as of March 2026. It publicly identifies individuals by name, country of origin, and alleged or actual criminal offenses. The narrow factual claim — that a database was launched — is accurate.

Framing as “win”: Deeply misleading. The claim presents the database as a straightforward transparency achievement. The evidence shows:

1. The database is a messaging tool, not a public safety tool. Internal ICE communications reveal the “Worst of the Worst” branding was developed as part of a coordinated viral social media strategy, with DHS hiring influencers and wedding videographers to produce “cinematic” content. The database exists to generate shareable content reinforcing the immigrant-crime narrative.

2. 56% of the people listed have no violent charges or convictions. The “Worst of the Worst” label is applied to a database where the majority of entries involve nonviolent offenses — drug possession, bad checks, traffic violations, shoplifting, immigration violations. The label is marketing, not description.

3. The database includes people merely charged, not convicted. Publicly listing people on a government website labeled “Worst of the Worst” based on unproven charges — without any mechanism for correction or removal — constitutes extrajudicial public shaming that undermines the presumption of innocence.

4. DHS admitted the database was rife with errors. Approximately 1,250 people were publicly listed with incorrect criminal charges. In at least one case, the White House publicly accused a man of a child sex crime he did not commit. DHS attributed this to a “glitch” and refused to explain what happened.

5. The database represents less than 5% of total ICE arrests. By the administration’s own implicit admission, 95% of the people ICE arrests do not qualify as “worst of the worst” — and even within the 5% that do, most have no violent charges.

What a reader should understand: Yes, the administration launched a publicly searchable database at WOW.DHS.GOV listing people ICE has arrested. That narrow fact is true. But the database’s actual contents contradict its branding: a majority of those listed have no violent criminal charges, the database includes people merely charged (not convicted), DHS admitted it contained hundreds of errors including at least one false accusation of a child sex crime, and internal ICE communications reveal the entire initiative was designed as a viral social media strategy rather than a public safety measure. The database functions as a digital pillory — publicly shaming noncitizens, including those not convicted of any crime, with no mechanism for error correction or removal. It is the visual component of the same narrative infrastructure that includes VOICE (Item #27): creating the impression that immigrants are dangerous criminals, despite the administration’s own data showing less than 14% of ICE arrests involve violent criminal records and 73.6% of detainees have no criminal conviction at all.

Cross-References

• Item #3: “650,000 arrests, detentions, and deportations” — the WOW database draws from this same pool of arrests; the database’s 30,000 entries represent less than 5% of the arrest population, most of whom had no violent criminal record
• Item #4: “400,000 criminal aliens deported” — the WOW database includes some of these individuals; both claims use the same “charged or convicted” formulation that conflates accusations with guilt
• Item #27: “Restored VOICE Office” — VOICE and the WOW database are complementary narrative tools; VOICE institutionalizes the premise that immigrant crime is a distinct threat, while WOW provides the visual evidence to make that premise feel immediate and visceral; both serve primarily as political messaging infrastructure

Sources

DHS. “DHS Unveils WOW.DHS.GOV: New Searchable Website Lets You Find Criminal Illegal Aliens Removed From Your Community.” December 8, 2025. https://www.dhs.gov/news/2025/12/08/dhs-unveils-wowdhsgov-new-searchable-website-lets-you-find-criminal-illegal-aliens

DHS. “DHS Offers Even Greater Transparency: Adds Another 5,000 Criminal Illegal Aliens to WOW.DHS.GOV.” December 18, 2025. https://www.dhs.gov/news/2025/12/18/dhs-offers-even-greater-transparency-adds-another-5000-criminal-illegal-aliens

DHS. “To Award the Mainstream Media for Raising Awareness About DHS’s Worst of the Worst Website, DHS Drops Another 5,000 Criminals, Now Featuring Over 30,000 Criminal Illegal Aliens.” February 20, 2026. https://www.dhs.gov/news/2026/02/20/award-mainstream-media-raising-awareness-about-dhss-worst-worst-website-dhs-drops

DHS. “DHS Recaps the Worst of the Worst Criminal Illegal Aliens ICE took Enforcement Action on During President Trump’s First Year in Office.” January 20, 2026. https://www.dhs.gov/news/2026/01/20/dhs-recaps-worst-worst-criminal-illegal-aliens-ice-took-enforcement-action-during

DHS. “WOW.DHS.GOV Website Makes it Easier to Search Your State.” January 22, 2026. https://www.dhs.gov/news/2026/01/22/wowdhsgov-website-makes-it-easier-search-your-state

Reason. “The Feds’ ‘Worst of the Worst’ Database Is Stuffed With Nonviolent Offenders. Who Exactly Is ICE Arresting?” December 12, 2025. https://reason.com/2025/12/12/the-feds-worst-of-the-worst-database-is-stuffed-with-nonviolent-offenders-who-exactly-is-ice-arresting/

Common Dreams. “New DHS Database Suggests That Less Than 5% of Those Arrested by ICE Are the ‘Worst of the Worst.’” December 2025. https://www.commondreams.org/news/dhs-worst-of-worst-database

CNN. “Exclusive: DHS Admits Its Website Showcasing the ‘Worst of the Worst’ Immigrants Was Rife With Errors.” February 19, 2026. https://www.cnn.com/2026/02/19/politics/homeland-security-worst-immigrants-website

The Washington Post. “We Read Thousands of Internal ICE Chats. They Show How Officials Make Raids Go Viral.” December 2025. https://www.washingtonpost.com/technology/interactive/2025/ice-social-media-blitz/

ProPublica. “We Found That More Than 170 U.S. Citizens Have Been Held by Immigration Agents.” October 2025. https://www.propublica.org/article/immigration-dhs-american-citizens-arrested-detained-against-will

CBS News. “Less Than 14% of Those Arrested by ICE in Trump’s First Year Had Violent Criminal Records.” January 18, 2026. https://www.cbsnews.com/news/ice-arrests-violent-criminal-records-trump-first-year/

Cato Institute. “5% of People Detained By ICE Have Violent Convictions, 73% No Convictions.” 2025. https://www.cato.org/blog/5-ice-detainees-have-violent-convictions-73-no-convictions

TRAC (Syracuse University). “Immigration Detention Quick Facts.” Data through February 7, 2026. https://tracreports.org/immigration/quickfacts/

Marouf, Fatma. “Immigration Law’s Missing Presumption.” Georgetown Law Journal. 2023. https://www.law.georgetown.edu/georgetown-law-journal/wp-content/uploads/sites/26/2023/07/GT-GGLJ230024.pdf

Mediaite. “DHS Admits Its Website Featuring ‘Worst of the Worst’ Arrests Is Full of Mistakes.” February 2026. https://www.mediaite.com/media/news/dhs-admits-its-website-featuring-worst-of-the-worst-arrests-is-full-of-mistakes/