Top Stories
All Sections
Finance
About
20 results for "hallucination"
A federal appeals court ruled that attorneys bear full professional responsibility for AI-generated errors in legal filings, closing off the AI-reliance defense and awarding attorney fees against sanctioned counsel.
A new survey finds eight in ten South Korean judges have presided over cases with AI-generated fake legal citations — and 92% are calling for fines, mandatory verification, and disciplinary action.
New York's court system issued guidance allowing attorneys to use AI in filings without disclosure while holding them fully accountable for any AI-generated errors, including hallucinated citations.
A U.S. Court of Federal Claims judge has ordered four government attorneys to explain why they should not be sanctioned for misrepresenting how the U.S. Army used AI to evaluate contract proposals — a case that goes beyond citation hallucinations into deliberate mischaracterization of AI's role.
US courts have escalated AI hallucination sanctions from a $5,000 landmark in 2023 to six-figure penalties in 2026, applying existing ethics rules to generative AI misuse — and the grace period is over.
The 10th Circuit Court of Appeals requires lawyers to certify AI-generated citations have been human-verified before filing — following sanctions in a Mississippi AI hallucination case.
Courts are sanctioning attorneys for AI-generated citations at a rising rate, while a review of 21 jurisdictions found no uniform standard — leaving lawyers personally liable under rules that vary dramatically by state.
A UK court issued a practice ban against a lawyer who submitted AI-hallucinated case citations, the first such disciplinary ruling in England and Wales — establishing that attorneys bear full professional responsibility for AI-generated content.
A Mississippi federal judge dismissed a case and sanctioned four lawyers after both sides submitted AI-generated filings they never reviewed — the most dramatic judicial response yet to AI hallucinations in courts.
An MIT study of 4.5 million federal civil cases found AI-written court filings jumped from 1% in 2023 to 18% in 2026, while self-filed docket activity surged 158% — threatening to overwhelm federal courts with hallucinated citations and fabricated legal quotes.
Oregon's Court of Appeals chief judge issued a public warning that AI-fabricated court filings are "rapidly escalating," with one attorney fined $10,000 and another $8,000 for AI-hallucinated citations.
Anthropic is moving directly into AI legal services, joining Harvey, EvenUp, and Legora as courts accelerate sanctions for AI hallucinations in legal filings.
AI hallucination incidents in legal filings are now documented at four to five new cases per day — and courts are considering mandating hyperlinked citations to catch fabricated sources at time of filing.
Courts are now sanctioning lawyers for AI-generated fake citations at four to five cases per day — a rate that has transformed a slow-building professional risk into an institutional emergency for law firms.
Alston & Bird's April 2026 AI Quarterly finds courts issuing consequential AI rulings without a coherent framework, including a privilege-destroying ruling for law firms using commercial AI tools.
A paper at ICLR 2026 finds that reinforcement-learning reasoning training — the technique behind o3 and Gemini Thinking — proportionally increases tool-call hallucinations as model performance improves.
Courts are now detecting four to five AI-hallucinated legal citations in filings per day, with sanctions exceeding $100,000—and law firms are adding clauses warning clients about attorney-client privilege risks.
Attorneys submitting AI-generated briefs with hallucinated case citations are facing mounting sanctions — while courts split on bans vs. deterrence, and law firms warn clients that chatbot use may destroy attorney-client privilege.
Major law firms are now writing AI privilege warnings into client contracts as courts sanction attorneys for AI-hallucinated citations — two converging risks the legal profession is only beginning to reckon with.
A federal court ordered an Oregon attorney to pay $109,700 for submitting AI-generated filings with fabricated case citations—the largest penalty yet for AI hallucination errors in a legal document.