For most of advertising’s modern history, realism was expensive. If you wanted a human face to sell something, you had to find a real person and put them on camera. That act alone forced a level of accountability on the production; you had to deal with contracts, likeness rights, and the looming shadow of reputational risk. Realism was the ultimate gatekeeper in this business because it required judgment. You didn't just pick a face; you stood behind a person.
AI lets us generate "synthetic humans" who look perfect, never complain, and cost next to nothing. But when realism becomes a cheap shortcut, we lose the friction that used to keep us honest. Shortcuts are where trust dies, because they blur the line between winning an argument and faking a connection.
New York’s new AI law isn't a tech ban, It’s now a reality check. Regulators aren't reacting to how cool AI is; they’re reacting to how easy it is to manufacture trust. The law is surgical. It doesn't care about your workflow or your post-production tools. It cares about one specific thing. Are you using a digital person to trick an audience into thinking they’re seeing a real human performance? If the answer is yes, you have to disclose it. Period. The state isn't judging your creativity, they’re judging your transparency.
The Price of a Wrong Call
As marketers, we’re used to optimizing for clicks and conversions. We often ignore the legal fine print until it’s too late. But starting June 9, 2026, the "ignore it" strategy gets expensive:
- $1,000 for the first violation.
- $5,000 for every violation after that.
Economically, a $5,000 fine is still less than hiring a live personality, but that’s the wrong way to look at it. If you’re running a programmatic campaign with hundreds of variations, these fines become a major financial liability. One mistake scaled across a digital buy could wipe out your entire margin before you even realize the campaign is live.
Why Judgment Matters More Than Automation
This law forces us to ask a question we’ve been avoiding: What does the audience actually believe they are seeing? That’s not a legal question; it’s a judgment question. In the past, being a spokesperson was a heavy responsibility. AI collapses that whole process into a single prompt. We’ve moved from “should we do this?” to “can we do this?” and that’s exactly where our judgment usually breaks. New York is simply making the choice for us. If it looks human, you have to be honest about it. Not because AI is "bad," but because you can’t build a brand on a lie.
The Big Picture
We have less than six months before the June 9 deadline. This isn't a "future" problem anymore; it's a Q2 creative problem. Think about the summer holidays. We’re going to see a flood of July 4th ads featuring "happy gatherings" and "backyard parties." Historically, these are high-volume socials that mimic real community gatherings. If those "people" are actually synthetic assets, will we see a shift back to real photography to avoid the disclosure label, or will brands lean into the "AI look"?
This law is not a constraint on creativity. It is a constraint on ambiguity. Used transparently, AI expands what brands can express. Used deceptively, it collapses trust faster than any bad campaign ever could. AI did not break advertising; undisclosed realism did. Production will continue to accelerate and creative possibility will expand, but trust is no longer a soft concept. New York just made it explicit.