Two senior national security officials tell me that there are more than a dozen secret and obscure watchlists that homeland security and the FBI are using to track protesters (both anti-ICE and pro-Palestinian), ā€œAntifa,ā€ and others who are promiscuously labeled ā€œdomestic terrorists.ā€

I can reveal for the first time that some of the secret lists and applications go by codenames like Bluekey, Grapevine, Hummingbird, Reaper, Sandcastle, Sienna, Slipstream, and Sparta (including the ominous sounding HEL-A and HEL-C reports generated by Sparta).

Some of these, like Hummingbird, were created to vet and track immigrants, in this case Afghans seeking to settle in the United States. Slipstream is a classified social media repository. Others are tools used to link people on the streets together, including collecting on friends and families who have nothing to do with any purported lawbreaking.

There’s practically nothing available that further describes what these watchlists do, how large they are, or what they entail.

ā€œWe came out of 9/11 with the notion that we would have a single ā€˜terrorist’ watchlist to eliminate confusion, duplication and avoid bad communications, but ever since January 6, not only have we expanded exponentially into purely domestic watchlisting, but we have also created a highly secretive and compartmented superstructure that few even understand,ā€ says a DHS attorney intimately familiar with the subject. The attorney spoke on the agreement that their identity not be disclosed.

Prior to 9/11, there were nine federal agencies that maintained 12 separate watchlists. Now, officially there are just three: a watchlist of 1.1 million international terrorists, a watchlist of more than 10,000 domestic terrorists maintained by the FBI, and a new watchlist of transnational criminals, built up to more than 85,000 over the past decade.

The new domestic-related watchlists—a set of databases and applications—exist inside and outside the FBI and are used by agencies like ICE and the Border Patrol to organize the Niagara of information in possession of the federal government. Collectively, they create ways to sort, analyze, and search information, a task that even artificial intelligence has failed to conquer (so far).

Among other functions, the new watchlists process tips, situation reports and collected photographs and video submitted by both the public and from agents in the field; they create a ā€œcommon operating pictureā€ in places like Minneapolis; they allow task forces to target individuals for surveillance and arrest; and they create the capacity for intelligence people to link individuals together through geographic proximity or what is labeled ā€œcall chainingā€ by processing telephone numbers, emails, and other contact information.

  • Basic Glitch@sh.itjust.worksOP
    link
    fedilink
    arrow-up
    41
    Ā·
    edit-2
    1 day ago

    When the CNN host pointed out that Pretti wasn’t violent, Blanche actually agreed, but went on to argue that there’s a third category for protest that is neither violent nor peaceful.

    Who doesn’t love a good ole fashioned American riddle?

    they create the capacity for intelligence people to link individuals together through geographic proximity or what is labeled ā€œcall chainingā€ by processing telephone numbers, emails, and other contact information.

    I feel like I’ve heard this story before, but where šŸ¤”

    The Lavender precedent

    Evidence recorded in the classified Israeli military database in May 2025 revealed that only 17% of the 53,000 Palestinians killed in Gaza were combatants. This implies that 83% were civilians

    Or, perhaps they were neither combatant or civilian, but some kind of 3rd category. Omg did I just solve America’s riddle?!?

    Lavender algorithm assigns each person a probabilistic score indicating the likelihood that they are members of an armed group. Six Israeli intelligence officers involved in the process testified that the system marked up to 37,000 Palestinians as suspected militants and identified their homes as targets for attack. These were predominantly low-ranking individuals.

    Indeed, they were called ā€œjunior operativesā€ in the military’s view, and would not have appeared on traditional target lists, which consisted of a handful of high-ranking officials.

    If an individual’s data patterns (such as in communications, travel, associations) resemble those of known militants, Lavender targets them for potential attack. However, even by the Israeli army’s own assessment, Lavender only sported a 90% accuracy rate. In other words, 1 in 10 people flagged by the machine were false positives by the IDF’s own submissions. These were civilians, protected from attack as per IHL, and yet killed as a result of algorithmic and command-and-control negligence.