AI Regulation & Policy

Today's Stories

1. 1

   Goldman, Morgan Stanley, Citi Demand Big Law Fee Cuts Over AI

   Goldman Sachs, Morgan Stanley and Citigroup are pressing elite law firms to lower fees, arguing AI is sharply reducing the cost of routine legal work such as research, document review and contract analysis. Citigroup has begun asking firms bidding for its business to disclose how much they are saving by using AI, and Morgan Stanley plans to put most of its outside legal work up for competitive bids by the end of the year. The billable-hour model has long made partners at top firms extraordinarily wealthy — top lawyers have "for a long time been compensated on the foundation of [associates billing for long hours]", as Morgan Stanley's general counsel Eric Grossman put it. The banks' push could amount to "a fundamental altering of the revenue foundation for these mega firms", Grossman added.

   Nearly half of large law firms surveyed this year by Citi's law firms group said AI had already affected how they price their work. Big Law's hourly charges have surged since 2023, with associate rates climbing 33% to an average $798 this year, according to Persuit data.

2. 2

   Music Publishers Sue Anthropic Over Lyrics Copyright

   Music publishers affiliated with Sony and Warner have filed a lawsuit against Anthropic, accusing it of copyright infringement on a scale described as one of the largest intellectual property thefts in history. The suit claims Anthropic obtained pirated copies of lyrics and scraped lyrics from the web. This case highlights the legal risks AI companies face when training models on copyrighted material without permission. The outcome could set a precedent for how AI firms use existing creative works, affecting the broader AI industry's approach to data sourcing.

   The lawsuit alleges both unauthorized acquisition of pirated content and systematic scraping of lyrics, which could lead to significant legal and financial consequences for Anthropic if the court rules against it.

3. 3

   Suginami Ward to Hire External Advisor for AI-era Fake News

   Suginami Ward, Tokyo, announced it will appoint one external advisor starting November to handle fake information on social media during disasters. The move responds to rising risks from generative AI. The advisor will train ward staff in information literacy and provide advice during incidents. The ward has budgeted 750,000 yen for the role, reflecting growing concern over misinformation spread.

   The advisor's term runs from November through March next year, with duties about five times a month. Candidates include former police officers and lawyers with crisis management expertise.

4. 4

   Anthropic opens Claude AI text detection to regulators, media, fact-checkers, and others

   Anthropic is launching its watermark verification API, letting approved organizations check whether text contains a digital watermark from Claude. The EU AI Act since August 2, 2025, requires new Claude models to embed invisible watermarks in their text output, making this detection system a compliance tool for regulators, law enforcement, media, fact-checkers, independent researchers, educational organizations, EU civil society groups, and enterprises.

   Anthropic plans to expand access over time. The system builds on Google's SynthID text method, tweaking word-selection randomness to create a statistically detectable pattern that, according to Anthropic, "may persist through some editing."

5. 5

   Pentagon adds ChatGPT, Grok to secure AI platform

   The Pentagon said on Monday it is adding military versions of ChatGPT and Grok to its secure AI platform GenAI.mil, which launched in December with only a military version of Google's Gemini. GenAI.mil already counts 1.7 million of the Department of War's 3 million-strong workforce. The move follows defense contracts worth up to $200 million each awarded to OpenAI and Elon Musk's SpaceXAI, which runs Grok. War Secretary Pete Hegseth has pushed for AI to become part of daily workflows, aligning with the White House's AI action plan.

   Service members can use both specialized versions for Controlled Unclassified Information (CUI) at Impact Level 5 (IL5). Notably absent is Anthropic's Claude, which had a falling out with the Department of War over its push to restrict use for autonomous weapons or surveillance.

6. 6

   New U.S. bill would tax AI firms to fund jobs if unemployment rises

   A new House proposal, introduced by Reps. Sara Jacobs, Greg Casar, and Valerie Foushee, would impose an excise tax on major AI companies. The tax rate would start at 2% on AI token value and 3% on AI service revenue when unemployment is 5% or less, and automatically rise as unemployment climbs. The bill is part of a broader bipartisan push in Congress to prepare for potential AI-driven job displacement. Similar efforts include a Senate proposal by Ron Wyden, a call from Elizabeth Warren to tax AI companies based on data center energy use, and Bernie Sanders' plan for a one-time 50% tax on OpenAI, Anthropic, and xAI.

   The bill's revenue would fund job creation in areas like housing construction, infrastructure, and child and elder care. No response yet from AI companies—OpenAI and Anthropic did not answer questions about supporting the proposed taxes.