As AI reshapes advertising, safety, and ethics take center stage. Will AI-driven decisions empower advertisers, or will big media control the rules?
Is the next chapter in the Advertising Safety or Brand Safety and Suitability conversations AI Safety?
Is AI Safety the Next Evolution of Brand Safety?
It would be hard to think about an article, keynote, or 2025 launch plan without the letters “AI” represented somewhere. How will we work with AI, what can it do, what do we want it to do?
It’s everywhere, and not to be the wet blanket—but in the world of innovation, there must be a conversation about security and safety–especially in advertising, where influence and outcomes are the North Stars.
What is AI safety when it comes to advertising?
What Is AI Safety in Advertising?
To consider the question, we must first truly consider how AI could assist advertising. I heard recently that the best way to think about AI is that you will have a machine in your pocket with a 130 IQ, with whom you can have conversations (and since they are connected to the internet), that knows everything about everything.
How might a company use this agent? Well, it could be a sounding board for ideas, surface analyzed data for decision-making, or make decisions on its own. And perhaps, most importantly for advertising, it might connect AI-generated ideas with AI-generated design to influence purchase decisions.
Three Big AI Safety Questions for Advertisers
This brings me back to safety, as this conversation is so new, we have to start from scratch and consider this scheme, so here are three framework ideas that might be a good place to begin:
If AI is surfacing data for analysis, how will we know the source of the data and the rights the AI has to use it?
How will we know if the creator of the AI hasn’t programmed it for purposes that influence people which are the opposite of your true goals or something externally nefarious?
Since the ad ecosystem is so complex, how will each member know if the other is following moral or business ethics?
These are just a few; I’m sure there are more…
Brand Safety Isn’t Always a Beloved Term—So Why Open This Door?
Brand Safety isn’t always a beloved term, so why would we open this door? For some context, it seems like Brand Safety is a concept that has felt like it tends to get in the way, and is a nuisance, like traditional compliance is sometimes considered.
Not surprisingly, the term has sometimes been conflated and substituted for censorship. I understand why—when you consider blocking anything, you might block the wrong things and overblock. When this happens, the entities on the wrong end get rightfully upset. That is just a fact, and this outcome has broad context for many disparate groups who all have an argument to make.
But I also know that an advertiser must be able to have control of their advertising targets and goals; that is also a fact.
Can’t both of those things be true? And how will AI enter into this part of the discussion?
The 2025 Debate: Who Controls AI in Advertising?
As I look out to 2025, here is the debate I see as Ad Safety with AI and traditional Brand Safety and Suitability continue to be re-imagined and re-defined. Will this technology truly be placed in the hands of every advertiser and democratized so it can be deployed for their own decision-making?
Or, will it be centralized behind a big media firewall to deploy on behalf of a marketer? Will ad adjacency be targeted or blocked by rules created by large language models, set by standards bodies, or potentially a series of community notes? Will there be new 3rd-party entrants with AI and LLM structures to deliver verification?
Which of these outcomes will lead to ad placements that make sense for all participants? Let’s see, but there is a lot at stake. 2025 is going to be interesting, to say the least.

Posted from https://www.admonsters.com/ai-safety-in-advertising-the-next-evolution-of-brand-suitability/
Comments