The US government is expanding its surveillance of social media to monitor millions of visitors and immigrants – and its embrace of more data analytics and artificial intelligence tools could increase scrutiny of US citizens as well.
“It is nearly – if not entirely – impossible for the government to focus only on non-citizens and not look at anyone else’s social media,” says at the Brennan Center for Justice, a public policy non-profit in New York.
Since 2019, the US Department of State has required the approximately 14 million people who apply for immigrant and non-immigrant visas each year to provide their social media information. But under the administration of US President Donald Trump, it has stepped up AI-enabled social media checks. For instance, the State Department began using AI tools to target student visa holders who express pro-Palestinian views as part of its new .
Advertisement
The State Department did not respond to questions about specific technologies used for social media monitoring but a State Department spokesperson says the agency will use “all available tools to receive and review concerning information”. The decision to revoke visas may also be based on “arrests, criminal convictions, and engaging in conduct that is inconsistent with the visa classification,” says the spokesperson.
The US Citizenship and Immigration Services (USCIS) – part of the Department of Homeland Security (DHS) – also recently it would consider “antisemitic activity on social media… as grounds for denying immigration benefit requests”. And the agency taking its surveillance further by collecting social media account names from anyone applying for immigration benefits, such as asylum or naturalised citizenship. If enacted, the policy would apply to millions of people each year.
Those millions of people could include US citizens as well as immigrants. For some people applying for full permanent residency, the policy would require the social media information of their “parents, stepparents or children, along with their spouse or former spouse, any of whom could be US citizens”, says Levinson-Waldman.
The public comment period for the USCIS proposal closes 5 May. But it is only one of many US government policies driving an increase in social media surveillance, enabled by new tech tools.
In its latest from April 2025, the DHS confirmed that its Customs and Border Protection agency has been using an AI-powered tool called Babel X to “compile social media and open-source information on travellers who may be subject to further screening”. The DHS and Babel Street, the company that developed Babel X, did not respond to requests for comment.
The Immigration and Customs Enforcement (ICE) agency, which focuses on detaining and deporting unauthorised immigrants under DHS, has also been using a social media monitoring tool called SocialNet that “aggregates data from over 200 sources, including Facebook, Twitter/X, Instagram and LinkedIn, as well as sites like dating apps and Roblox”, says Levinson-Waldman.
Social media posts are frequently filled with “irony, memes, algo speak and all this other indecipherable stuff” that can be easily misunderstood by people, says at the Center for Democracy & Technology, a public policy non-profit in Washington DC. “And now we’re layering on top of that the use of AI that is error prone.”
ShadowDragon, the company that developed SocialNet, says the program does not incorporate AI. “We built our tools intentionally to prevent dragnet-level surveillance or [disclosure of] private data that is inaccessible to the general public – and we don’t believe in trusting machines to make decisions about what the data means,” says at ShadowDragon. “We share many of the concerns from the folks you’ve already talked to, which is why there’s no artificial intelligence, machine learning, or data analysis technology in our products. Our tools simply automate the gathering of public information for people responsible for making their own decisions about the data.”
More than 11,000 ICE agents have also been using tech tools provided by the company LexisNexis Risk Solutions for “screening, vetting and targeting for deportation as of 2022”, says Levinson-Waldman. Meanwhile, the company’s describes its ongoing ICE contract as promoting public safety and not being used to prevent legal immigration. “We have supported public safety efforts across multiple administrations for more than 30 years, regardless of political leadership,” says a LexisNexis Risk Solutions spokesperson.
It appears that the US federal government is investing more public funds in such tech tools, says at Just Futures Law, an immigration advocacy non-profit in Washington DC. “We’re witnessing a real-time expansion of the use of social media monitoring technologies,” says Shah. “When you use social media monitoring to intimidate, harass, alienate, deport, incarcerate, arrest – when that becomes your standard to do those things – it’s antithetical to a lot of what democracy stands for.”
