Back to articles
DevPickView feed
Hacker NewsHN submitter: doodlesdev
Original published:

Brazilian Court Sanctions Lawyers for Prompt Injection in Court Filing

Brazilian Court Sanctions Lawyers for Prompt Injection in Court Filing
AI summary
Editorial policy

A Brazilian court fined two lawyers $16,000 after their AI detected white-on-white prompt injection instructions hidden in a court filing.

Key points

  • •Two lawyers hid instructions in a court filing attempting to manipulate the court's AI to favor their case.
  • •The prompt injection was concealed using white text on a white background but was detected by the court's AI system.
  • •The court sanctioned the lawyers with a fine of $16,000 for the violation.
Read the original(opens in a new tab)
#Security & Misuse#Regulation, Ethics & Society

Related articles(Security & Misuse)

See more
INTERNET Watch

Apple to begin checking 'harmful content' in iCloud public links [Yajiuma Watch]

Apple announced the introduction of a feature to check for harmful content in public iCloud links.

Hacker News

Cuban: Facebook knowingly runs AI deepfake scam ads; docs say $16B fraud revenue

Cuban alleges that Facebook knowingly uses his face in AI-generated scam ads for nonexistent diabetes cures, with documents indicating $16 billion in fraud revenue.

ITmedia

OpenAI Announces Cybersecurity-Specialized 'GPT-6 Cyber', Discovers 81 Unknown Vulnerabilities in Firefox

OpenAI announced the cybersecurity-specialized AI 'GPT-6 Cyber' for defenders, aiming to support rapid protection and modification of critical infrastructure.

Hacker News

Claude Code sends your email address to websites

A bug in Claude Code was reported where it leaks email addresses in User-Agent headers during web searches, causing users to receive numerous scam emails.

Hacker News

Introducing Pangram Image 1.0

Pangram Image 1.0, a new AI tool for detecting images and videos, has been introduced.