Full Breakdown
Congressional Staff’s AI-Generated Amendment Summary Triggers Debate Over Legislative Drafting Practices
6/25/2026, 3:02:44 AM
AI-Generated Amendment Summary Sparks Controversy
A defense-bill amendment summary drafted by Representative Anna Paulina Luna’s (R-FL) office was posted online, revealing a timestamp, filler text and the label “Claude responded:” that identified Anthropic’s Claude AI as the source. The unedited document was submitted to the House, sparking scrutiny of AI use in legislative drafting.
Background: AI Tools in Congressional Work
Luna confirmed her staff used AI to correct a draft and submitted it without edits. On X she wrote, “Not a shocker. Most staff use it.” The sources did not cite any formal congressional policy on AI use.
Key Actors
- Anna Paulina Luna – Representative (R-FL)
- Claude – Anthropic AI chatbot
- Grok – Alternative AI model
- Max Steele – Senior director, Everytown for Gun Safety
- Zaid Jilani – Journalist
- Andrew Egger – White House correspondent for The Bulwark, who posted a photo and highlighted Luna’s response.
Data & Statistics
The amendment summary that circulated included a timestamp, garbled filler text, and the explicit label “Claude responded:” indicating the text was produced by Anthropic’s Claude AI without human editing.
Official Statements & Responses
Luna’s office acknowledged the oversight and said staff must “double-check and be more thorough” in future submissions. The statement framed the incident as an isolated lapse and affirmed Luna’s appreciation for AI tools alongside a call for tighter editorial review.
Criticism & Opposition
Max Steele of Everytown for Gun Safety called the error “not a mistake,” labeling it intentional incompetence. Journalist Zaid Jilani urged, “Let’s ban congressional staff from using AI to draft legislation.” Both called for stronger oversight of AI in lawmaking.
Why It Matters
The episode underscores concerns that AI-generated text can be introduced into official legislative documents without adequate review, raising questions about accuracy, accountability and public confidence in the lawmaking process.
Verbatim Quotes
- “Yeah, my staff used AI to correct a draft text and didn't edit,” — Anna Paulina Luna, Representative (R-FL)
- “Not a shocker. Most staff use it.” — Anna Paulina Luna, Representative (R-FL)
- “Btw love Claude but Grok is way more savage,” — Anna Paulina Luna, Representative (R-FL)
- “It wasn't a mistake. We were being incompetent intentionally, actually.” — Max Steele, Senior Director of Communications, Everytown for Gun Safety
- “Let's ban congressional staff from using AI to draft legislation.” — Zaid Jilani, Journalist
Conflicting Reports & Gaps
The sources do not provide an official congressional policy on AI-generated text, nor do they disclose whether any formal investigation has been launched. Consequently, the extent of institutional oversight remains unclear.
What’s Next
Advocates now urge explicit rules for AI use by congressional staff, ranging from voluntary guidelines to legislative bans. The controversy could shape future policy discussions on AI oversight in Congress.
