Again, Noem cites false data. Approximately 60% of people in ICE custody have no criminal history whatsoever.
@margyoh
Immigration Law and Policy Expert | Senior Fellow, Brennan Center for Justice at NYU | Migration and Human Rights Fellow, Cornell Law School | Former Biden Administration Domestic Policy Council | Former Department of Justice
Again, Noem cites false data. Approximately 60% of people in ICE custody have no criminal history whatsoever.
Last week, Lyons said 200k kids were lost-today Noem says upped it to 400k. Neither is true. Kids were released to vetted sponsors, not lost. DHS wants to deport them, which
Noem herself conceded, although she said they were “reuniting” kids with family outside the US.
During her Senate testimony, Sec Noem claims that many immigrant criminals have come to the U.S., but immigrants commit crimes at lower rates than people born in the US, and communities with high concentrations of immigrants have lower rates of crime. www.brennancenter.org/our-work/ana...
The SAVE Act could block millions of Americans from voting. It is not common sense. If passed, it would be the first time in history Congress passed a vote suppression law. #SOTU
The SAVE Act is an attack on the freedom to vote that would block millions of American citizens from voting. #SOTU #SOTU2026
U.S. elections are secure and accurate, but now the federal government is threatening that infrastructure. www.brennancenter.org/our-work/res...
25/ Congress should also use its power to close the data broker loophole and strengthen privacy protections against AI-enabled surveillance. It should demand transparency about how the military is using AI in hostilities – a critical first step towards regulating the technology.
24/ Congress has the constitutional power and duty to regulate the military. It should reassert its authority over matters of war and peace.
23/ This appears designed to pressure Anthropic to waive its usage restrictions on Claude – restrictions that might reflect not only the company’s own policy preferences but also what U.S. and international law require.
22/ DOD has threatened to designate the company as a supply chain risk, which would not just cut it off from future military work but also chill its business with other companies that themselves have a stake in defense contracting.
www.axios.com/2026/02/16/a...
21/ Anthropic is no doubt aware – and perhaps increasingly wary – of the legal issues created by the Pentagon’s use of Claude in military operations. But relying on Anthropic – or any company – to restrain the military is a fool’s errand.
20/ All questions Congress must demand the Pentagon answer in its investigations of military actions in Venezuela, on the high seas, and elsewhere.
www.nytimes.com/2025/12/02/o...
19/ Was Claude, Maven or any other AI used to distinguish strike targets from civilian infrastructure, assess the civilian impact of strike options, or otherwise plan the attack?
For that matter, how is the military using AI in its boat strikes, and prep for possible strikes on Iran?
18/ The Venezuela attack struck several apartment buildings and killed two civilian air traffic controllers in a car leaving the Caracas international airport.
www.newyorker.com/news/the-led...
16/ Given that Congress did not authorize the Venezuela operation, and that Venezuela did not pose an imminent threat to the US, the attack was unconstitutional on its face. This makes any use of AI to facilitate the attack unlawful.
www.brennancenter.org/our-work/ana...
15/ Remember the public uproar in 2018 over Google’s involvement in Maven, the Pentagon’s AI pilot to analyze troves of drone and satellite footage? It’s now a full-fledged program the military has credited with speeding up targeting.
breakingdefense.com/2025/05/ai-u...
13/ Then there’s the use of AI to “assist weapons targeting without sufficient human oversight.” All attacks – AI-assisted or not – must comply with U.S. law and the international law of war.
11/ AI enables DOD to take this data and connect the dots with other discrete datasets about US persons to recreate their movements, associations, and political views at scale – an invasion of privacy far more egregious than the type of surveillance before the Carpenter court.
10/ But law enforcement and intelligence agencies - and DOD - have bought their way around this protection, spending millions to purchase entire databases of Americans’ location information and other sensitive data from data brokers on the commercial market.
int.nyt.com/data/documen...
8/ What is clear is that AI-enabled spying raises serious Fourth Amendment concerns that neither Congress nor the courts have directly grappled with – in large part because the law is not keeping up with how quickly the technology is evolving. But that doesn’t mean DOD has a blank check.
7/ Why would the military be spying on Americans? When it is monitoring targets overseas, for example, it likely sweeps up Americans’ e-mails, phone records, location information and other personal data - all of this is just as likely to route through Latin America or the Middle East as the U.S.
6/ But are the disputed uses of Claude lawful? Using the model to “spy on Americans” can mean many things – from running it across multiple data sources to generate personal information about Americans to trying to reconstruct this information from the data the model was trained on.
5/ Defense secretary Pete Hegseth has declared, “We will not employ AI models that won’t allow you to fight wars.” And DOD insists that it should be free to use Claude however it wants, as long as it is obeying the law.
media.defense.gov/2026/Jan/12/...
4/ This rift is emerging at a time where the Pentagon is pushing for breakneck adoption of AI in warfighting.
www.war.gov/News/News-St...
3/ In January, Anthropic and DOD were deadlocked over the appropriate uses of Claude. Anthropic wants assurances that the Pentagon will not use Claude to “spy on Americans or assist weapons targeting without sufficient human oversight,” which would violate the company’s usage policies.
2/ Anthropic signed a $200 million contract last year to develop “frontier AI capabilities” for DOD. Its AI model, Claude, has now been integrated into classified systems developed by Palantir and Amazon for the military.
www.semafor.com/article/02/1...
1/ The dispute between Anthropic and DOD over the limits of using AI in warfare is escalating. I’ll have more to say on what these limits should be in a @brennancenter.org report out next month. But for now, here’s what you need to know, and what’s truly at stake 🧵:
www.nytimes.com/2026/02/18/t...
It is also critical to remember in this debate that immigrants commit crimes at lower rates than people born in the US, and communities with high concentrations of immigrants have lower rates of crime. www.brennancenter.org/our-work/ana...
The remaining 35% have either a pending charge (but no conviction), a misdemeanor immigration conviction for “improper entry,” or a conviction for another nonviolent offense.
Bumping this up because Senator Lankford also used incorrect numbers. To repeat: based on ICE’s own data, 60% of those arrested and detained have NO criminal history, and only 5% have a conviction for a violent offense.