Ask someone on the street what AI actually is, and the answers spread out fast.
"It's a really, really smart computer that does extremely crazy things out of your imagination," one Bay Area resident told CBS San Francisco. Another called it exciting. A third said it mostly makes her nervous about her job. None of them are wrong exactly. They're just describing different pieces of a term that keeps moving.
A Definition That Keeps Moving
Even the people who've spent decades studying AI don't have a clean answer. Chris Brooks, a computer science professor at the University of San Francisco who's worked in the field for more than 30 years, put it simply.
"It's a shifting definition. It seems like every time we solve something, the target moves," Brooks said.
That's not a new phenomenon. The first AI workshop happened at Dartmouth in 1956. Deep Blue beat Garry Kasparov at chess in the late 1990s. For years afterward, progress plateaued in ways that made AI feel more like an academic curiosity than a consumer product. Then, over roughly the last decade, and especially the last three or four years, the technology moved from research labs into everyday tools people use without necessarily realizing it.
Michele Neitz, founding director of USF's Center for Law, Tech, and Social Good, describes AI as an umbrella term covering machine learning, deep learning, neural networks, and whatever comes next, rather than one single thing.
Why the Ambiguity Matters More Than It Seems
That fuzziness might sound like an academic problem. It isn't, not anymore.
Federal regulators have decided they don't need a precise definition of AI to go after companies that misrepresent what their products actually do. The FTC's Operation AI Comply, launched in 2024, has continued bringing cases against businesses that exaggerate or fabricate AI capabilities in their marketing, and the agency opened more than a dozen new cases last year alone.
The SEC has moved in parallel, treating misleading AI claims made to investors as a potential securities law violation, not just a marketing exaggeration.
The two agencies are working from existing law, not new AI-specific statutes. That matters because it means companies can't wait for clearer AI regulation before taking marketing claims seriously. The FTC and SEC are applying decades-old fraud and deception rules to AI claims right now, treating an overstated "AI-powered" label the same way they'd treat any other false statement about what a product does.
The FTC Isn't Waiting for Consensus
The enforcement has real teeth. DoNotPay paid a $193,000 settlement after marketing its chatbot as "the world's first robot lawyer" without evidence to back the claim.
More recent cases have expanded who's liable. Under what regulators call the means and instrumentalities doctrine, the FTC has started charging not just the company making AI claims directly to customers, but the vendors who supplied the deceptive marketing materials in the first place. A substantiation standard that once applied only to consumer-facing claims now applies equally when the buyer is another business.
FTC Chairman Andrew Ferguson has framed the crackdown as pro-innovation rather than anti-technology, arguing in testimony before Congress that the market only functions if businesses and consumers can actually trust what companies say their AI does. That framing matters for how seriously companies should take it. This isn't regulators being hostile to AI adoption. It's regulators applying ordinary truth-in-advertising standards to a category of claims that's been unusually easy to inflate.
What This Means for Miami
Plenty of Miami businesses, from fintech startups to real estate platforms to law firms, market themselves as AI-powered in some form. Given the current enforcement climate, that phrase now carries real legal exposure if it isn't backed by something a regulator could actually verify.
The practical fix isn't complicated. Any Miami company using "AI-powered," "machine learning-driven," or similar language in marketing, investor materials, or sales conversations should be able to document specifically what the AI does and how, the same way it would substantiate any other factual claim about its product. Given regulators aren't waiting for the industry to settle on a shared definition of AI before enforcing against it, Miami businesses shouldn't wait either.
