In other words, the question of where to draw the line is inseparable from the question of how much money marketers are prepared to lose by being honest.

“How will marketers react to this, knowing that there is a lot of anxiety in the market right now around AI and its impact on society and the economy,” said Nada Bradbury, CEO of AD-ID.

The industry is trying to figure it out. Industry groups are developing guidance, brands are drawing their own lines and regulators are lobbying for something better than the blunt instruments currently being considered. But the process is slow and the clock is not. The European Union’s AI transparency regulations expire in August. Starting in June, New York state will require disclosure of AI-generated humans in marketing. Lines that the industry hasn’t yet drawn are about to be drawn by people with very different priorities.

“I think in the last few months alone, we’ve had five, six, seven brands asking us, “Do you have any guidance?” “Can we get some insight into how brands are thinking about this issue?” said Gabriel Robitaille, policy director and AI community lead at the World Federation of Advertisers.

The reason, she said, is the same thing that drives most compliance conversations: proximity. As August approaches, legal teams that used to leave questions to marketers are now looking for answers, and marketers are looking for a place to turn. Anyone who has experienced the arrival of the General Data Protection Regulation will understand the feeling. The question now is: What exactly should marketers do with the time they have left?

WFA’s answer is intentionally narrow. Robitaille said the starting point is that the basic principles of advertising self-regulation still apply, regardless of what technology is used. Do not use it to mislead. Do not use to generate unsubstantiated claims or exaggerate the effectiveness of a product. She argued that once you remove all of these potentially harmful use cases, the gray area that remains is actually very small.

What remains is a question of thresholds. WFA’s position is that labeling should begin when AI is central to the commercial nature of the message, i.e. when it materially shapes what someone believes about the product.

Below that line, disclosure requires judgment. The responsible thing to do then is to label it. The most obvious case is synthetic humans, Robitaille said, and the public’s fears are consistent and well-supported across studies. The beach background generated behind the shampoo bottle is another matter entirely. Although it is technically AI, it is unlikely to influence someone’s purchasing decisions.

WFA is not alone in trying to solve this problem. In January, the IAB announced what it called the industry’s first uniform standard for AI disclosure in advertising, and it arrived at much the same destination. Don’t label everything. Focus on content that can truly deceive you. Treat synthetic humans as the most obvious trigger. Background changes, audio enhancements, and post-production tweaks all fall below what the IAB calls “standard production techniques.”

Where it goes further is in the details. Digital twins of living people depicted in fabricated events need to be labeled in the same way as prompt-driven AI images and videos. Disclosures are required for deaths rendered by AI, and the IAB also requires that all ads be accompanied by C2PA metadata. So platforms like Meta, TikTok, and YouTube can see exactly what AI was used for and apply their own labels if necessary.

“Advertiser response to this framework has been thoughtful and pragmatic,” said Caroline Giegerich, vice president of AI at IAB. Most brands are not trying to hide their use of AI, but are instead seeking clarity. The real question brands are grappling with is not, “Should I disclose?” But when does disclosure provide a meaningful service to consumers, and when does it cause confusion or fatigue? It’s understandable that there is a sensitivity at this point to classifying something as being generated by AI. This term can have unintended connotations ranging from fraud to manipulation. ”

When the term AI carries such connotations, some brands are drawing the obvious conclusion that they should not be associated with AI at all. Enter the “No AI” disclaimer – the certified organic label for your ads.

Take Aerie, an underwear brand owned by American Eagle, for example. The company used Pamela Anderson in its ads to promote a pledge it had already made last October not to use AI-generated bodies or people. In it, the actor is shown prompting the chatbot to create a model, before revealing that the model was a real human all along. The promise itself is an extension of the brand’s 2014 pledge not to retouch characters in ads. For Aerie, “not using AI is not a compliance decision, it’s in its brand DNA.

Baby products brand Coterie went even further, publicly pledging to no longer use AI-generated images in its social media marketing. The company’s CEO, Jess Jacobs, told the Wall Street Journal that the move is a confidence strategy in a crowded market with a crowded audience of viewers who are particularly skeptical of parents. Le Creuset goes out of its way to make it clear that its latest social content (a visually original video by digital artist Ian Padgham that transforms the brand’s signature cookware into unlikely objects) has no AI involved. Before anyone thinks otherwise, an explanation will be given in the comments of those posts, unadvertised.

This is a remarkable change for an industry that has spent three years telling customers that AI will change everything.

“Over time, consistent transparency will normalize the role of AI in advertising, just as we have normalized other technologies in the past,” Giegerich said. “The industry’s goal should not be to over- or under-label, but to disclose information in a way that provides information without being overwhelming and builds trust rather than suspicion.”

Numbers you need to know

$30 billion: Anthropic’s reported run-rate revenue

$102 billion: OpenAI predicts total advertising revenue by 2030

61%: Percentage of US retail decision makers use media mix modeling to measure incrementality.

78%: Percentage of US Millennials (non-Gen Z) likely to use a second screen during a 2026 World Cup match.

What we covered

Who is David Dugan, Global Head of Advertising at OpenAI?

We knew Fidji Simo, CEO of Applications, had been looking for a head of advertising since September last year. Currently, David Dugan has been appointed to this role. Digiday spoke to a number of people who know him to understand why OpenAI thought he was the right person for the job.

TikTok’s Cartoon Weiss to exit platform as head of global operations in North America

TikTok’s other top executive, Khartoon Weiss, is leaving the platform today after nearly six years with the business, according to a memo seen by Digiday.

Trade Desk is changing how advertisers buy and what they can see

The Trade Desk is testing two new automated buying modes within the Kokai platform that streamline campaign management and combine media, data and technology fees into a single price.

OpenAI’s TBPN contract reveals the limits of branded entertainment

The value of TBPN lies in the fact that Meta CEO Mark Zuckerberg, who rarely gives coverage, sat down with the host. That’s Microsoft’s boss, Satya Nadella. Altman himself keeps coming back.

what we are reading

Why Anthropic’s new model upsets cybersecurity experts

According to the platform, Anthropic has effectively said that this model is too dangerous to release to the public, as a model of this caliber could soon be accessible to criminals, hackers, and nation states.

OpenAI asks California, Delaware to investigate Musk’s ‘anticompetitive conduct’ ahead of April trial

According to CNBC, OpenAI sent a letter to the attorneys general of California and Delaware urging them to investigate anti-competitive conduct by Elon Musk and his associates.

Why acquiring a podcast is actually the first step to building an AI ad stack

AI Ad Economy details how OpenAI’s acquisition of TBPN supports the AI ​​platform’s advertising ambitions.

Perplexity’s monthly revenue increases by 50% due to pivot from search to AI agents

The AI ​​platform’s annual recurring revenue is estimated to have increased to more than $450 million last month as it launched new agent tools and moved to usage-based pricing, according to the Financial Times.