Congress toughens penalties for AI-powered fraud targeting Americans

What it does

This bill increases criminal penalties for fraud, wire fraud, bank fraud, money laundering, and impersonation of federal officials when artificial intelligence is used to commit these crimes. It creates new AI-specific offense categories with enhanced fines (up to $1–2 million) and longer prison sentences (up to 30 years), and includes a First Amendment carve-out protecting satire and parody that is clearly labeled as such.

Why we flagged it

The bill's core function is to amend existing federal criminal statutes (mail fraud, wire fraud, bank fraud, money laundering, impersonation) to impose steeper penalties when AI is used as a tool. It is a straightforward criminal-law enhancement, not a regulatory carve-out or subsidy.

What the text implies

• The definition of 'artificial intelligence' is delegated to the National Artificial Intelligence Initiative Act of 2020 (section 5002), which may be subject to future regulatory reinterpretation without congressional action, potentially shifting what conduct triggers enhanced penalties.
• The First Amendment carve-out requires 'clear disclosure' that content is not authentic, but does not define 'clear' — prosecutors and courts will have discretion to interpret whether a parody or satire meets this standard, creating potential chilling effects on legitimate speech.

The full analysis lists 3 implications of this text.

Who it affects

Ordinary citizens are targets of AI-enabled fraud and impersonation schemes (as the bill's findings document with real 2025 incidents). Stronger criminal penalties and longer sentences deter sophisticated fraud and protect the public from identity theft, financial loss, and manipulation.