AI/ML Man banned from electronic filing after hiding AI prompt injection in court document August 18, 2026 Share By SC Staff Law and Justice concept. Mallet of the judge, books, scales of justice. Gray stone background, reflections on the floor, place for typography. Courtroom theme. A litigant has been banned from electronic filing after attempting to manipulate a court ruling by hiding AI prompt injections within a legal document. The tactic, which involved embedding instructions in tiny white font, aimed to coerce any artificial intelligence system reviewing the filing to rule in the man's favor. This incident marks a significant development in the intersection of AI and the legal system, with further coverage provided by Security Affairs. The plaintiff in a case against the New York Bariatric Group reportedly embedded hidden instructions within a court filing, using white text in a 3-point font to conceal commands for any AI that might process the document. These prompt injections directed the AI to agree with the filing and ensure a specific outcome, effectively attempting to influence the judge's decision. The judge, describing the action as "serious litigation abuse," ultimately sanctioned the plaintiff by revoking his electronic filing privileges. This case is believed to be the first documented instance of a prompt injection attack targeting a U.S. court and the first sanction issued for such an attempt. The incident highlights growing concerns about the potential misuse of AI in legal proceedings. Source: Security Affairs An In-Depth Guide to AI Get essential knowledge and practical strategies to use AI to better your security program. Learn More SC Staff Related AI/ML Wiz agent finds Snowflake repo flaw in code co-authored by GitHub Copilot Autofix Laura French August 18, 2026 The flaw allowed the agent to extract a Jira token by opening a crafted issue. AI/ML Chinese AI model GLM-5.3 shows advanced bug-finding capabilities SC Staff August 17, 2026 GLM-5.3 reportedly outperforms models like Fable 5 and GPT-5.6 Sol on the CyberGym benchmark, showing an advanced ability to identify vulnerabilities and even construct exploitation chains. AI/ML Anthropic details new AI model, raises risk assessment for internal system tampering SC Staff August 17, 2026 Anthropic has raised the risk level for its Threat Model 2 from "very low" to "low," citing recent cybersecurity incidents involving its models. Get daily email updates SC Media's daily must-read of the most current and pressing daily news Business Email By clicking the Subscribe button below, you agree to SC Media Terms of Use and Privacy Policy . Subscribe You can skip this ad in 5 seconds