ICE Is Reading 8 Billion Posts a Day

Your tweets are being analyzed for deportation leads. Here's how.

TL;DR: In September 2025, ICE signed a $5.7 million, five-year contract with Zignal Labs for an AI system that analyzes over 8 billion social media posts daily. It extracts faces, geolocations, and metadata from your public posts and generates "curated detection feeds", watchlists for enforcement. The same company provides "tactical intelligence" to the Israeli military operating in Gaza. ICE also hired 30 contractors to monitor Facebook, Instagram, TikTok, X, and YouTube around the clock. Your public posts are now immigration enforcement data.

The $5.7 Million Deal

In September 2025, ICE's Homeland Security Investigations unit signed a five-year contract with Zignal Labs through government reseller Carahsoft Technology [1].

Contract details:

  • Value: $5.7 million
  • Duration: Through September 2028
  • Purpose: "Real-time data analysis for criminal investigations"
  • Buyer: ICE Homeland Security Investigations intelligence unit

That's government-speak for: we're using AI to turn your social media into enforcement leads.

8 Billion Posts Per Day

Zignal Labs doesn't just monitor social media. It inhales it.

The system processes:

  • Over 8 billion social media posts daily
  • Content in more than 100 languages
  • Posts from Facebook, Instagram, TikTok, X (Twitter), YouTube, and more

What the AI extracts:

  • Metadata (time, location, device information)
  • Geolocation data embedded in posts
  • Faces using computer vision
  • Logos and text using OCR
  • Social connections between accounts

Your vacation photo with geotags? Analyzed. Your profile picture? Face-scanned. Your follower list? Mapped for connections.

"Curated Detection Feeds", AKA Watchlists

Here's where it gets concerning. Zignal generates what they call "curated detection feeds", automated lists of accounts flagged by the AI for investigators [2].

How it works:

  1. AI scans billions of posts for keywords, patterns, or connections
  2. System flags accounts matching certain criteria
  3. Flagged accounts appear in "detection feeds"
  4. ICE investigators review feeds for enforcement targets

Critics call these what they are: watchlists generated by algorithms with no transparency about how they decide who gets flagged.

The Surveillance Technology Oversight Project (STOP) described the deal as "an assault on democracy and free speech" [3].

From PR Firms to Gaza

Zignal Labs didn't start as a surveillance company. They pivoted.

Company timeline:

  • 2011: Founded in Silicon Valley
  • Original clients: PR firms and political campaigns
  • Original purpose: Media monitoring and narrative tracking
  • 2021: Formal pivot to defense and intelligence
  • 2021: Created "public sector advisory board" with national security veterans

Current government clients:

  • US Secret Service (since 2019)
  • US Marines
  • US State Department
  • Department of Defense
  • Israeli Defense Forces

A 2025 Zignal Labs pamphlet highlighted their work providing "tactical intelligence" to "operators on the ground" in Gaza [4]. The same AI analyzing your Instagram is feeding intelligence to military operations.

30 Contractors Watching Around the Clock

Zignal is the AI. But ICE also hired humans to watch your posts.

According to Wired, ICE plans to employ nearly 30 contractors dedicated to social media monitoring [5]:

  • 12 contractors based in Vermont
  • 16 contractors based in California
  • Some required to be "available at all times"
  • Monitoring: Facebook, Instagram, TikTok, X, YouTube

Their job: Turn posts, photos, and profiles into enforcement "leads" for ICE's Enforcement and Removal Operations Division.

That's a room full of people whose entire job is watching social media for deportation targets. 24/7.

Zignal Isn't Alone

Zignal Labs is just the newest addition to ICE's social media surveillance arsenal:

Tool Vendor Capability
Zignal Labs Zignal Labs 8B posts/day, AI analysis, detection feeds
Babel X Babel Street Social media monitoring across platforms
ShadowDragon ShadowDragon Maps online activity, links profiles to identities
Giant Oak Giant Oak Social media vetting for visa applicants

Together, these tools create a comprehensive social media surveillance system. Your posts are being analyzed by multiple AIs from multiple angles.

What the ACLU Says

Patrick Toomey, deputy director of the ACLU's National Security Project, was blunt [6]:

"[The Department of Homeland Security] should not be buying surveillance tools that scrape our social media posts off the internet and then use AI to scrutinize our online speech."

Toomey added: "Agencies shouldn't be deploying this kind of black box technology in secret without any accountability."

The ACLU's concerns:

  • No public independent audits of the system
  • No clear policies on data retention
  • No transparency about who gets flagged and why
  • No accountability when the AI makes mistakes

The Chilling Effect

When people know their posts are being monitored for enforcement, they stop posting. That's the point.

EFF and Yale Law School attorneys noted that social media surveillance tools "exacerbate the chilling impact" on free expression [7].

Who self-censors:

  • Immigrants afraid to post about their lives
  • Activists afraid to organize online
  • Family members afraid to mention undocumented relatives
  • Anyone who doesn't want their face in a government database

A group of labor unions sued over the program, calling it a "mass, viewpoint-driven surveillance program" that targets immigrants for their political speech [8].

The lawsuit is ongoing. The surveillance continues.

Protecting Yourself

If your posts are public, they're being analyzed. Here's what you can do:

Basic steps:

  • Make accounts private: Won't stop everything, but reduces exposure
  • Disable geolocation: Turn off location tagging on all posts
  • Audit your photos: Faces are being scanned with computer vision
  • Review your connections: Social graphs are being mapped
  • Consider what you post: Assume enforcement agencies are reading

For higher-risk individuals:

  • Use pseudonymous accounts not linked to real identity
  • Avoid posting photos with identifiable faces
  • Use VPN to mask IP addresses
  • Consider whether you need to be on these platforms at all
  • See our Social Media Privacy Hardening guide

Remember: Private accounts can still be accessed through legal process or security breaches. The only truly private post is the one you don't make.

The Bottom Line

ICE is using AI to analyze 8 billion social media posts every day. They're extracting faces, locations, connections, and metadata. They're generating watchlists. They've hired 30 people to watch your feeds around the clock.

The company providing this capability also provides "tactical intelligence" to military operations in Gaza. They started as a PR monitoring firm and pivoted to government surveillance when the money got good.

There are no public audits. No clear policies. No accountability when the AI flags the wrong person.

Your public social media posts are now immigration enforcement data. That's not speculation. That's a $5.7 million government contract.

References

  1. Jacobin: ICE Will Use AI to Surveil Social Media (October 2025)
  2. VisaVerge: ICE Invests Millions in Zignal Labs Social Media Surveillance AI (2025)
  3. Truthout: ICE Just Spent Millions on a Social Media Surveillance AI Program (2025)
  4. Wikipedia: Zignal Labs
  5. Wired: ICE plans to hire 30 contractors for social media monitoring (October 2025)
  6. The Lever: ICE Just Bought A Social Media Surveillance Bot (2025)
  7. EFF: Social Media Surveillance and Free Speech (2025)
  8. Business & Human Rights Resource Centre: ICE launches AI-powered surveillance system (2025)