Senators voted overwhelmingly in the early morning hours yesterday to drop the proposed pause on state AI laws that would have dissuaded state lawmakers from regulating artificial intelligence at the local level for the next five years. This latest move opens the door for states like California, New York, Colorado, Illinois, and more to regulate the use of AI in the workplace. Yesterday’s 99-1 vote will clear a path for more laws to be passed in states across the country and almost certainly lead to compliance headaches for multistate employers given the impending regulatory patchwork. What should you do to prepare?
Quick Background
- Original Proposal: The House passed a sweeping proposal to impose a 10-year moratorium on state-level laws regulating AI on May 22 as part of the budget package. Read our summary here.
- Senate Turns Ban into Pause: Senate leaders veered and introduced a revised proposal on June 5 that would have merely blocked states from receiving federal tech funding if they regulated AI in the next decade. Read our summary here.
- Trimmed to 5 Years: Negotiations over the past weekend led the Senate to trim the pause window to five years, and also created exceptions allowing states to regulate copyright issues and children’s safety. Read our summary here.
What Happened?
Shortly after 4:00 this morning, the Senate voted 99-1 (with retiring Sen. Thom Tillis (R-NC) the lone no-vote) to completely jettison the AI pause from the budget bill. The quick reversal is a bit of a surprise given that the measure was supported by:
- AI Czar David Sacks
- White House technology adviser Michael Kratsios
- Commerce Secretary Howard Lutnick
- Senate Commerce Chair Sen. Ted Cruz (R-TX)
- Silicon Valley tech leaders
What’s Next?
While the tech industry will certainly lobby Congress to consider future legislative proposals to ease regulation on the use of AI, attention now turns to the states – where efforts are well underway to build guardrails and compliance hurdles that employers will need to navigate. You can read our most recent summary here recapping all existing state laws, all pending laws set to take effect in the coming months, and those key proposals brewing in statehouses across the country that will soon be voted on by state lawmakers.
What Should You Do?
1. Audit Now, Not Later
Employers should audit AI tools for explainability, bias, and disparate impact (regardless of the recent executive order taking on that legal theory) and not wait for regulators to catch up. If your vendor can’t explain how their system works or prove it’s compliant with civil rights laws, it’s time to rethink that partnership. Here are the essential questions you should ask your AI vendor before deploying AI at your organization.
2. AI Governance is the Name of the Game
Risk assessments, transparency, and human oversight remain essential tools for preventing AI-based discrimination. Read AI Governance 101: The First 10 Steps Your Business Should Take in order to plot out the best course for your organization.
3. Prepare for the Patchwork
As we expect a slew of new state regulations to roll out over the next few years, design your compliance systems with flexibility in mind. Read our most recent summary here.
