Federal judges confront growing gap between law and technology
Two U.S. federal judges acknowledged this week that laws written decades ago cannot keep pace with the speed of technological change, creating an expanding legal gray zone as courts grapple with AI-related cases and privacy disputes.
U.S. District Court for the District of Columbia Chief Judge James Boasberg and U.S. District Court for the District of Massachusetts Judge Allison Burroughs discussed the challenge at the IAPP Global Summit 2026 in Washington, D.C., as their digital-related caseloads grow more complex.
Burroughs pointed to existing Fourth Amendment protections against unreasonable searches, rooted in constitutional language from 250 years ago. Courts have stretched old statutes-like the pen register law, which governed phone metadata-to cover modern cell tower data collection. The question now is whether existing laws can be retrofitted or if Congress must write new ones.
"The gap is getting bigger for two reasons," Burroughs said. "One is that there's so much more data stored electronically, and the other is that there are just so many ways of gaining access to data."
AI in the courtroom raises immediate problems
Boasberg recently presided over a case where lawyers filed briefs containing fabricated citations generated by AI. He ordered that side to pay the opposing counsel's attorney fees as a sanction.
Despite not using AI himself, Boasberg said judges cannot ignore the technology. AI tools have already appeared in federal judicial opinions, sometimes embarrassingly so when those opinions contained hallucinated citations.
"The question is how can it help without compromising privacy issues, sealed cases; there's just a whole lot that we have to figure out," he said.
Legal research tools already integrate AI
The main legal research platforms-LexisNexis and Westlaw-have integrated AI features. However, the versions available to judges have those features disabled. Several judges in the U.S. Court of Appeals for the First Circuit are testing the AI capabilities in a pilot program.
Burroughs sees narrow uses for AI in judicial work. The technology could handle repetitive administrative tasks-the boilerplate text judges cut and paste between cases. But AI cannot replicate the judgment calls required in opinion writing.
"The things that AI are not going to be able to capture are judgment issues, not a matter of reciting boilerplate text," Burroughs said.
Could AI decide cases someday?
Boasberg raised a hypothetical: if an AI model trained on repeated case types achieved 95% accuracy, would litigants prefer a quick AI decision over waiting years for a judge to rule?
"I'd feel lucky if I were 95% accurate," he said.
Administrative hearings-Social Security appeals, immigration cases-might be the first place AI plays a decision-making role, given the massive backlogs in those systems.
Disclosure becomes the immediate fix
For now, Burroughs recommended judges require lawyers to disclose when they use AI in filings. Lawyers must certify they have checked citations to catch hallucinations.
"They can use AI, they can't use AI, they must disclose when they're using it," she said. "It's hard to think of what these rules would be going forward today."
Courts are moving faster than legislation. Without clear rules, judges are writing them on the bench, case by case.
For professionals working in law, understanding AI for Legal applications and the risks of Generative AI and LLM tools is becoming essential as these technologies enter courtrooms and legal practice.
Your membership also unlocks:
