What You Post on Social Media Can Be Used Against Your Immigration Case. USCIS Is Using AI to Look.
USCIS is using artificial intelligence to screen social media accounts as part of immigration case review
CBP officers at airports and ports of entry are looking at phones and social media profiles. What you post, what you are tagged in, who you are photographed with, and what your accounts say about your relationships and your life can all be used as evidence in your immigration case — or against it.
I had a client whose visa application was flagged because a photo on their social media showed what an AI tool identified as a gang sign. It was a peace sign. That case had to be explained and documented at length. It cost time and money and stress that was entirely preventable. The system is not always accurate and the consequences of its errors fall on the applicant.
This is one of the most underappreciated risks in immigration law right now. I want to explain what is happening, what it means for different types of cases, and what you should do about your accounts today.
What USCIS and CBP are actually doing
USCIS has been conducting social media screening since approximately 2019. The program has expanded significantly and now uses AI tools to scan publicly available social media content for information relevant to immigration applications. Officers are looking for inconsistencies between what an applicant stated on their application and what their social media suggests about their life, relationships, criminal associations, national security concerns, and fraud indicators.
CBP — Customs and Border Protection — has separate authority at ports of entry, including international airports. They can ask to see your phone. They can ask for your social media handles. In some cases they have demanded passwords, though the legal authority for that remains contested. The AI screening expands the government's capacity to review accounts at a scale that human officers could not manage alone.
What this means practically is that content you posted years ago, content you were tagged in by someone else, and content that is technically public even if you think of your account as private are all potentially in scope for review.
What they are looking for
The most common issues I see flagged are relationship inconsistencies in marriage-based cases, gang or criminal association indicators, and statements or content that contradict something on an application.
In marriage-based green card cases, your social media is essentially evidence about your relationship. Officers look at whether your accounts show a shared life. Are you and your spouse tagged together in photos? Do your posts reflect the relationship timeline you described on your application? Is your spouse labeled as your spouse on your accounts? Is your relationship history on Facebook accurate and consistent with what you submitted? If you claimed you married in 2022 but your Facebook shows a relationship status change in 2021 or 2023, that inconsistency gets flagged and has to be explained.
Gang affiliation — or anything that could be interpreted as gang affiliation — is another area where AI screening has produced false positives. Hand gestures in photos, tattoos, neighborhood associations, people standing in the background of images, phrases in captions that have dual meanings in different contexts. The AI does not have context. It identifies patterns and flags them. A human then reviews the flag, but the flag creates a problem that has to be resolved before your case can move forward.
Statements that contradict your application also come up regularly. Claiming you have never been affiliated with a certain organization and then having photos from that organization's events on your account. A post about traveling to a country you said you had not visited. A check-in at a location that contradicts your stated residence. A comment that suggests a relationship you did not disclose. These inconsistencies are exactly what the AI screening is designed to surface.
What to do about your accounts right now
If you have a pending immigration application or you are planning to file one, go through your social media accounts before you submit anything.
Make your accounts private. This does not prevent CBP from requesting access at a port of entry, but it limits what AI tools can scrape from public searches and reduces the surface area of what can be reviewed without direct access to your device.
Label your spouse correctly on every platform. If you are married and filing a marriage-based green card, your Facebook relationship status, your Instagram bio, and how you tag your spouse in posts are all part of the evidence. They should consistently reflect reality.
Review photos you are tagged in by other people. You cannot control what others post, but you can remove tags from photos that could be misread — hand gestures, events, people in the background whose associations you cannot account for. If a photo could be misinterpreted out of context, removing your tag removes it from your public profile.
Do not delete everything at once. Bulk deletion of content right before an immigration application or interview can itself raise questions about what you were removing and why. Selective, thoughtful review is better than a mass deletion that leaves an obvious gap in your account history.
Be careful about who you are connected to publicly. Associations in your friends list, in your comments, in your tags — connections with people who have criminal histories or gang affiliations can be flagged even if you personally have nothing to do with their activities.
Do not post about your immigration case. No updates about applications, hearings, approvals, or denials. No comments about your status. No content that gives context about your case that you have not already disclosed on your application. What you say publicly about your own case can be used at your interview.
For marriage-based green card cases specifically
Your social media is part of your case file in the eyes of an immigration officer. It should tell the same story as your application. Before your green card interview, review both your application and your social media to make sure they are consistent. An officer can and does look at accounts.
I have seen cases where officers brought printed screenshots from a client's social media to the green card interview. Not necessarily to catch anyone in a lie — just as part of the review process. If what they printed matches what you said on your application, the interview moves forward normally. If it does not, you are explaining an inconsistency across a table from an officer with the power to deny your case.
The bottom line
Social media is not private in immigration proceedings. Treat every post, photo, tag, and connection as potentially visible to an immigration officer. That is not paranoia — that is the current reality of how these cases are reviewed.
If you have concerns about your social media and a pending immigration case, or if something on your accounts could be misread and you are not sure how to handle it, call us before your interview or before you file. A consultation can walk through what needs to be addressed.
Attorney Arzoo Connor
ARC Legal Services | Fort Worth, TX | Hablamos Español
469-200-0158 | arclawoffice.com
This post is for general informational purposes only and is not legal advice. Every immigration case is different. Please consult a qualified attorney about your individual situation.












