AI Week in Review - 5/17/25

The Details

Federal

AI and the labor market: A scenario-based approach
Federal Reserve Governor Michael Barr outlines two scenarios for AI's impact on labor—gradual augmentation or radical transformation—highlighting implications for employment, inequality, and future monetary policy strategy.

US Copyright Office releases GenAI training guidance
The U.S. Copyright Office’s report explores whether using copyrighted works to train generative AI requires consent or compensation, highlighting legal uncertainty, fair use debates, and potential licensing frameworks.

Republicans move to freeze state AI rules
A new Republican tax bill seeks to block states from regulating AI for a decade, prioritizing innovation over oversight and aligning with Trump’s deregulatory push amid growing state-level AI legislation.

HHS to do more with help of AI
Health Secretary Robert F. Kennedy Jr. told Congress his department is aggressively adopting AI to cut costs, boost efficiency, and revolutionize clinical trials and drug approvals.

FBI issues warning about AI voice impersonations of US officials
The FBI warns that AI-generated voice messages impersonating senior U.S. officials are being used in targeted phishing attacks to access sensitive data from current and former government personnel.

NCIS, FBI officials see promise in AI use cases
NCIS and FBI officials highlighted successful AI deployments in HR and cybersecurity triage, showing how federal law enforcement is using AI to improve back-office efficiency amid budget and workforce constraints.

Commerce rescinds Biden-era AI export controls
The Commerce Department has rescinded the Biden-era AI export rule, citing innovation and diplomacy concerns, and will issue new guidance to prevent U.S. tech from aiding adversaries like China.

Air Force launches AI Center of Excellence to speed up AI adoption
The U.S. Air Force is launching an AI Center of Excellence to accelerate AI adoption, streamline prototype-to-mission transitions, and align AI capabilities with defense priorities.

State / Local

Here’s how Arizona schools are using AI to rethink education
Two Arizona schools are pioneering AI-integrated education models that personalize learning, free up teachers for deeper engagement, and emphasize real-world skills like entrepreneurship and financial literacy.

States chase OpenAI’s $100 billion AI American Dream
States across the U.S. are fiercely competing to host OpenAI’s $100B Stargate project, despite rising concerns over data center impacts, limited long-term jobs, and costly tax incentives

OpenAI governance opens it to state AG pressure
OpenAI’s nonprofit governance structure invites ongoing scrutiny from state attorneys general, particularly in California, raising questions about how its mission to serve humanity will be enforced.

Scaling GenAI in state and local government
GenAI is driving early success in U.S. state and local governments—from streamlining archives to cutting translation times—but scaling remains the key to unlocking its full potential.

International

The global rise of public at
Governments worldwide are increasingly building public AI infrastructure, adopting varied models that reflect national priorities and signal a strategic shift away from market-driven AI development.

Ireland releases guidelines for responsible use of AI in the public service
Ireland has released detailed guidelines to ensure AI use in the public service is responsible, human-centered, and aligned with public trust and rights through structured tools and principles.

Is AI the future of America's foreign policy?
CSIS's Futures Lab is testing AI models in diplomacy, revealing biases and unpredictable behaviors as researchers explore using AI to assist in high-stakes peace negotiations and crisis decision-making.

House of Lords pushes back against government’s AI plans
The House of Lords backed an amendment requiring AI firms to disclose copyrighted training data, challenging government plans that would allow use without permission and sparking a broader clash over creators’ rights.

Saudi crown prince launches new company to develop AI technologies
Saudi Arabia launched “Humain,” a state-backed AI firm under Crown Prince Mohammed bin Salman, aiming to position the kingdom as a global AI hub and diversify its economy beyond oil.

Development of GenAI from a copyright perspective
A new EU study highlights the need for tailored copyright solutions and licensing frameworks to support both generative AI innovation and intellectual property rights protection.

U.S. and UAE partner to build massive AI data center
The U.S. and UAE will co-develop a 5-gigawatt AI data center complex in Abu Dhabi—America’s largest such deployment abroad—advancing tech ties and global AI infrastructure.

UK AI to revolutionize public engagement
The UK government is rolling out an AI tool, "Consult", to analyze public consultation responses up to 1,000 times faster than humans, aiming to save £20 million annually and free up staff time.

Everything Else

5 ways to instantly level up your communication using AI tools
Executive coach Andrea Wojnicki shares five advanced ways professionals can use AI tools to enhance communication—from brainstorming and metaphors to tone optimization and audience feedback.

Pope identifies AI as a main challenge for humanity
Pope Leo XIV named AI a key global concern in his inaugural address, signaling continuity with Pope Francis while emphasizing his unique spiritual and pastoral priorities.

Something alarming is happening to the job market
Recent data shows rising unemployment among college graduates, signaling that AI and broader economic shifts are eroding traditional white-collar career pathways for young professionals.

OpenAI launches new AI coding agent
OpenAI has launched a research preview of Codex, a new AI coding agent that autonomously writes, debugs, and reviews code—signaling a major push into AI-driven software engineering.

Meta delays "Behemoth" AI model release
Meta has delayed the release of its largest Llama 4 AI model, “Behemoth,” amid concerns it doesn’t significantly outperform earlier versions—signaling a broader industry reckoning with AI scaling limits.

AI is the greatest threat—and defense—in cyber
AI is rapidly transforming cybersecurity, simultaneously escalating threats like phishing and deepfakes while empowering defenders with advanced detection, automation, and agentic AI capabilities.