AI went viral among attorneys. We have the numbers on what happened next

Uptake and the perfect storm

Generative AI swept into law firms fast — faster than many expected. It has been reported that attorneys embraced models for drafting pleadings, depositions and memos because the output looks polished, professional and, crucially, complete. That’s the rub. These systems are brilliant at producing structured legal prose that reads like an expert wrote it, and equally brilliant at inventing perfectly plausible but false case citations. The result is a perfect storm: powerful productivity promises colliding with professional duties that hinge on ironclad accuracy.

Fake cases slipped through

The hazard showed up in court filings. The Southern District of New York case from 2023 is the poster child, but it was only the beginning. It has been reported that courts and bar associations began tracking a rise in filings containing AI-generated “cases” and citations that don’t exist, and some of those allegedly slipped past routine fact-checking and reached judges. Lawyers, of course, are officers of the court; they have an ethical duty to verify facts. When AI makes it easy to look right while being wrong, the emotional hit is sharp — the public’s trust in the adversarial system takes a hit, and the professional embarrassment is real.

Fallout and the response

Where does this leave the profession? Regulators, courts and firms are scrambling. It has been reported that some jurisdictions are tightening disclosure rules and threatening sanctions; bar associations are advising attorneys to disclose AI use and to verify outputs rigorously. Training, checklists, and automated verification tools are the low‑hanging fruit. But structural fixes will be harder: can you inoculate a system against plausible nonsense without killing the very efficiency lawyers sought? The legal system is built on precedent and proof — and when the building blocks start to look counterfeit, everyone has to lean in and fix it.

Sources: The Register