Overzealous regulation won’t help AI policy keep pace with innovation

When was the last time you used facial recognition to unlock your phone? How about a digital assistant like Siri or Alexa to streamline tasks with simple voice commands? Or a meeting transcription service that freed you from taking notes during calls? Or an online shopping platform that automatically identified the best deals among thousands of listings?

Everyday tools powered by artificial intelligence have saved you time, money and hassle. They’re also equipping small business owners that stand up the backbone of our economy to thrive in crowded digital marketplaces. So why is California poised to adopt overzealous regulations at odds with Gov. Gavin Newsom’s 2023 executive order committed to shaping “the future of ethical, transparent and trustworthy AI, while remaining the world’s AI leader”?

Despite a once-promising trajectory for AI innovation in California, the state’s current process threatens to do more harm than good, with murky proposals from both its legislature and ever-growing web of government agencies. While state legislators consider approximately 50 AI-related bills, the California Privacy Protection Agency (CPPA) – an agency governed by appointees rather than elected representatives – is moving to formalize its own AI regulations. Many conflict with proposed legislation, as well as Newsom’s executive order directing the state to guide, as opposed to control, the continued rollout of AI technologies.

The more complicated the legal roadmap, the more time and resources it takes to follow its directions. Larger companies may have the resources to navigate these twists and turns, while small business owners are disproportionately left behind. It is only fair that laws are clearly written and not overly burdensome to comply with. It is only fair that laws are associated with reasonable penalties for noncompliance. Unfortunately, the suite of AI proposals before California lawmakers are neither fair nor logical.

Assembly Bill 2930, introduced by Asm. Rebecca Bauer-Kahan, which covers “automated decision tools,” would fine businesses $10,000 per violation, while allowing the state attorney general’s office to bring additional civil penalties of as much as $25,000 per violation. The governor, Legislature and regulatory bodies should be on the same page before laws are enacted. It makes no sense for elected leaders and an appointed board to pursue competing points of view about a subject as complicated and consequential as AI.

The issue with the CPPA is its attempt to regulate the use of AI across all industry sectors, although the agency’s purview is data privacy, specifically. The agency’s jurisdiction is hard to define because the technology it seeks to regulate – like automated decision-making (or ADM), where a computer computes a decision, as opposed to a human deciding one – is already a key part of so many everyday functions: automated bill payments and direct deposits, voice assistants, social media algorithms, automated email sorting and filtering, colleges’ admissions decisions and more. Any regulations on ADM end up applying to almost every industry.

This regulatory overreach threatens to stifle innovation and diminish the vibrant digital economy that is a hallmark of our state. We ask legislators to step up and rein in the power of the CPPA with sensible checks and oversight before it adopts misguided rules that trample ingenuity. We urge leaders of this regulatory agency to work collaboratively with Newsom and the Legislature, in consultation with academics and issue experts, on crafting AI guardrails that position California to be the world’s AI leader.

Let’s protect businesses, employees and taxpayers from the well-documented dangers of technology gone rogue while continuing to support tools that save people time, money and hassle.

Tracy Hernandez is Founding CEO of the Los Angeles County Business Federation,