A bipartisan group of Senate leaders unveiled a long-awaited artificial intelligence “roadmap” yesterday, laying out their hopes for potential legislation that would govern the use of AI across the country and ensure the U.S. stays on the frontline of AI innovation. And while the roadmap covers several key workplace-related proposals, what’s not included in the roadmap is almost as important – meaning we can expect to see a great deal of AI governance left to the states. What are the 10 things employers need to know?
1. This isn’t a new law – or even a bill
The roadmap is the culmination of almost a year’s worth of work by Senate leaders. We covered the AI Insight Forums held by the Senate last year as they laid the groundwork for the policy proposals we see in yesterday’s roadmap. However, the roadmap – crafted by Sen. Chuck Schumer (D-N.Y.), Sen. Mike Rounds (R-S.D.), Sen. Martin Heinrich (D-N.M.), and Sen. Todd Young (R-Ind.) – doesn’t include any specific legislative language. It reads more like a wish list of proposals they would like to see become reality.
2. Workplace implications are a big part of the roadmap
While the 31-page roadmap covers a broad spectrum of issues, the role of AI in the workplace plays a prominent role. It specifically calls for employers, unions, and employment-related organizations to be involved in the legislative development process given all of the many issues arising.
3. Job displacement is a concern
The roadmap notes that many workers are worried about AI impacting their jobs – from blue-collar positions all the way up to C-suite executives. It calls for the exploration of possible solutions – including private sector best practices – to the impact that AI will have on the long-term future of work. It also calls for the development of an “appropriate policy framework” in response, including ways that government can help combat disruptive job displacement.
4. Incentives to upskill workers
The roadmap also calls for the development of legislation that would boost training, retraining, and upskilling workers so they can successfully participate in an AI-enabled economy. It specifically mentions the idea of offering incentives for businesses that develop strategies to integrate new technologies and reskill employees.
5. Proposed immigration reform to help support AI growth
It also recommends that lawmakers consider legislation to improve the immigration system to allow more highly skilled STEM workers into the county, which would foster advances in AI technology development in the U.S.
6. Roadmap calls for data privacy push
The roadmap also promotes a robust approach to data privacy in the AI era. “The legislation should address issues related to data minimization, data security, consumer data rights, consent and disclosure, and data brokers,” according to yesterday’s announcement. Interestingly, however, the roadmap does not specifically endorse the proposed comprehensive data privacy bill sitting in Congress (the American Privacy Rights Act) – which would require employers to notify applicants and workers when AI is used for workplace decisions and also allow workers to “opt out” of such use.
7. What’s NOT included is almost as interesting
But interestingly, the roadmap does not promote the kind of workplace-related AI measures that have started to spring up in states like Colorado, California, and New York. Nowhere in the roadmap do we see a proposal requiring AI bias audits, nor any sort of anti-discrimination of bias protection. While several states have already proposed a mandatory notice requirement for employers who use AI for job-related purposes, the roadmap steers clear of any such proposal. Some employers might view this as good news, while others may worry about yet another area where we’re likely to now see a patchwork of varying state laws that complicate life for multistate operations.
That said, there is acknowledgment of existing concern around the potential for disparate impact and unintended harmful bias. The “High Impact Uses of AI” section in the roadmap states: “when any Senate committee is evaluating the impact of AI or considering legislation in the AI space, the AI Working Group encourages committees to explore how AI may affect some parts of our population differently, both positively and negatively.”
8. Roadmap covers a variety of non-employment topics
You may be interested in the non-workplace proposals contained in the roadmap (which you can find here). Among them:
- Support for AI research and development at the private level
- Deepfake identification and prevention (especially related to election content)
- Increased spending for national security efforts
- Identifying ways to ensure higher education institutions can compete in AI innovation
- Reviewing reports from the U.S. Copyright Office and U.S. Patent and Trademark Office on how AI impacts intellectual property law
9. Expect to see various separate bills instead of one massive package
The Senate leaders who crafted the roadmap indicated that they don’t want to see a massive package of intertwined legislation, or even one big piece of legislation. Instead, you can expect to see a series of individual bills tackling various aspects of the roadmap. The priority will be on regulating deepfakes, especially ahead of the 2024 elections.
10. There’s little chance most of these suggestions will become law in 2024
The reality, however, is that these proposals stand little chance of becoming law in 2024. The combination of a historically fractured Congress, reticence by Republicans to regulate industry too closely and inhibit innovation, and the fact that time is already running out on the legislative calendar will all conspire against any sort of speedy progress. Senator Schumer announced that he soon plans to meet with House leadership to begin working together to develop the bills that would advance the goals of the roadmap, but 2025 seems a better timetable for most of the workplace-related proposals.
