Technology has transformed the practice of immigration law and continues to do so. Practitioners from the past few decades will remember collecting forms from the office of United States Immigration and Naturalization Service and typing information on a typewriter. Some forms were required in multiple copies, with sheets of carbon in between. Many forms were required to be filed on specific colours of paper, including blue, yellow, green and pumpkin. If we did not want to mail in the forms, we could walk into the government office, pay the fee and file the forms in person.

Today, we use forms programs or file forms online using credit cards or fund transfers (no paper cheques allowed). Since COVID, electronic reproductions of forms with wet signatures on them have been accepted for submission. While some filings are still done in person at ports of entry, appointments may be required and may be established online. No colour-coded forms are required.

So, what is the next major technological shift in the practice of immigration law?

AI is fundamentally changing the way practitioners think about, strategize and plan for, and prepare both their cases and their clients. Not only are practitioners using AI in their research, submissions and writing, but clients are using it as well. Seasoned lawyers with law degrees, licences to practice and decades of practice are being challenged in their assertions and are having their writing adjusted by clients’ AI usage. Caution should be exercised.

AI can speed up research and has many positive uses, such as screening for typographical or grammatical errors. And since a lawyer’s time is a client’s money, opportunities for savings abound. However, AI can be, and not infrequently is, either materially misleading or flat-out wrong. It sometimes asserts an answer that is directly contrary to the truth. Or it may assess an article and reach a conclusion that is contrary to that of the author. AI can also make up great sounding resources, including statutes, regulations and policies, that may or may not exist. It may also include attributed quotes that sound great but do not actually exist. In our experience, it has even referenced actual laws, policy manuals or organizational authorities, while providing completely fake quotes as to what is in the resource in order to give the answer it believes the “researcher” is looking for. This presents a huge danger to laypersons and seasoned immigration practitioners alike, who may not go far enough down the rabbit hole to ask AI, “Is that a real [quote/law/case], or did you make that up?”

However, government officials are on the prowl to identify AI-written or AI-adjusted letters and evidence submitted in support of immigration cases. AI-generated documents and language use prose that often sounds stilted, flowery or exaggerated, along with certain grammatical markers — such as excessive em dashes (like the ones used around this phrase) — that can help officers and AI-detection software employed by the government easily identify AI-generated arguments, attestations and evidence. Where this occurs, it can lead to severe consequences, ranging from casting doubt on the truthfulness of assertions or evidence to denials and possibly even charges of fraud, creating long-term inadmissibility (for individuals filing their own paperwork) or disbarment or ethics sanctions (for immigration practitioners).

The U.S. government is using AI in other ways too. It can detect if a signature is a scan of a wet signature or a copied signature affixed to a form. If signatures are copied or digitally inserted signatures rather than scans of wet signatures, this can result in denial of a file before it even gets to the merits-based analysis of eligibility. Government-use AI can also detect the metadata used in creating a form, providing the adjudicator information on prior uses of a template. AI itself can be utilized to determine whether AI was used in the creation of a document, although it is unclear whether that enhances or diminishes its credibility. AI can also be used to generate boilerplate requests for additional evidence (RFEs) that may be lengthy, inaccurate or unrelated to the specific eligibility criteria for a particular benefit, or that may request documentation that is not probative. This leaves individuals or practitioners with a short window (usually 30 days) to figure out how to answer these requests.

Other technologies create different types of problems or solutions for immigration practitioners or individuals filing their own cases. Most of these issues stem from the fact that technology does not always work perfectly. Glitches in government software, as well as private immigration-related software and systems for generating forms, can cause problems with submissions to the government and payments for filings. Some individuals are concerned about the security of their information and documents, both within their lawyers’ offices and systems and within government-utilized systems for storing and vetting information. History suggests that these concerns are legitimate, as hostile actors are continually attempting to hack both private-industry information systems and government databases.

Technology and its use in immigration law are not going away and are likely to continue advancing at an increasingly rapid pace. This will continue to change how the government interacts with individuals and their attorneys, and there may be errors and glitches that could cause upheaval and chaos. These errors and glitches can significantly impact the lives of foreign nationals and their ability to study, work and live in the U.S. both now and into the future.

The best way to protect yourself is to exercise caution when using technology in any meaningful way. AI-generated “research” needs a deep level of vetting before you should trust or use any element of the information it provides. All work utilizing technologies, AI or otherwise, should be proofread and carefully reviewed before being submitted to the government to ensure information is accurate and evidence is real. While technology has made information easier to access, it remains an unreliable resource for vetting the truthfulness of available information or providing accurate and reliable information every time. When using technology, the old adage “trust but verify” is a necessary mindset for anyone seeking to navigate the immigration process successfully.

As technology continues to evolve the practice of immigration law, the attorneys at Lippes Mathias LLP are closely monitoring its impact on our clients. If you have questions about how technology may affect your immigration matter, or would like to discuss another business immigration matter, please contact Elizabeth M. Klarin (eklarin@lippes.com), Eileen M. Martin (emartin@lippes.com) or any member of our Immigration practice team.

The opinions expressed are those of the author(s) and do not necessarily reflect the views of the author’s firm, its clients, LexisNexis Canada, Law360 Canada or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.