There's a version of the AI-in-marketing conversation that I keep hearing at conferences and on LinkedIn that goes something like this: AI is coming for marketing jobs, agencies should be scared, everything is changing overnight. And then there's what's actually happening, which is more interesting and more specific than the panic suggests.
AI tools are getting good at the tasks that take a lot of time but not a lot of judgment. Writing first drafts of ad copy. Pulling together competitive research. Building audience segments from behavioral data. Generating creative variations for testing. Summarizing campaign performance across platforms. These are real capabilities, and they're getting better fast. But they're all execution tasks. They're the "what" and the "how." The strategic thinking, the "why" and the "for whom," still requires a person who understands the business, the market, and the customer.
Here's where it gets uncomfortable for a lot of agencies and internal teams: if AI can do in 20 minutes what used to take your team a week, and the output is comparable, then the value your team was providing was labor, not strategy. That's fine if the pricing reflected it. But a lot of agencies charge strategy-level fees for execution-level work, and AI is about to make that gap very visible.
The transparency effect
Having worked inside and alongside agencies for most of my career, the honest truth is that a lot of what gets billed as "strategy" is really just organized execution. Pulling a media plan together. Compiling a reporting deck. Running the keyword research. Setting up audiences. These tasks require skill, but they're repeatable processes that AI can accelerate dramatically.
When a business owner or marketing director starts using AI tools themselves, even casually, they begin to understand what's hard and what isn't. They realize that generating 50 headline variations isn't a two-day project. They realize that a competitive analysis doesn't take 40 hours. They start asking their agency or internal team better questions, and those questions often don't have satisfying answers.
"Why did this report take a week when I can get something similar from an AI tool in an hour?" That's a fair question. And the agencies that can answer it with "because we added these specific strategic insights that changed your allocation and saved you $30K last quarter" will be fine. The ones that can't answer it have a problem that no amount of AI adoption on their end will solve.
What AI actually changes
The biggest shift I see isn't that AI replaces people. It's that AI compresses the time between question and answer. Before, if you wanted to know how your campaigns performed across channels, you waited for a reporting cycle. Someone pulled data from five platforms, organized it, made charts, and presented it two weeks later. By then, the window to act on most of those insights had closed.
Now, AI can synthesize cross-platform data in near real time. It can flag anomalies, suggest allocation shifts, and generate insights that would have taken a skilled analyst a full day. This is genuinely useful. But the person reading that output still needs to know what to do with it. Should you shift budget from Meta to Google because the AI flagged a CPA increase? Maybe. Depends on whether that increase is seasonal, whether you're in a testing phase, whether there's a creative fatigue issue, or whether the attribution model is even trustworthy. That judgment call is where the value lives.
AI also changes how fast you can test creative. Instead of producing three ad concepts and running them for a month, you can produce 30 and start getting signal in a week. But somebody still needs to design the test, interpret the results, and decide what to do next. The speed of execution went up. The need for strategic thinking didn't go down.
AI compresses the time between question and answer. But knowing which questions to ask, and what to do with the answers, is still the hard part. That's where experienced marketers earn their value.
The agency model under pressure
For agencies specifically, AI creates an uncomfortable question about pricing. If you've been billing 15 hours a month for reporting and AI cuts that to 3 hours, do you lower the fee? Do you reinvest those 12 hours into deeper strategy work? Or do you pocket the efficiency and hope the client doesn't notice?
The agencies I respect are being upfront about this. They're saying: "AI makes us faster at the operational stuff, and we're reinvesting that time into strategic work that actually moves your business." They're showing clients what they're doing with the extra capacity. More creative testing. Deeper competitive analysis. Proactive recommendations instead of reactive reporting.
The agencies that will struggle are the ones that built their entire value proposition around doing tasks that AI now does cheaper and faster. If your pitch is "we manage your Google Ads campaigns" and that's the whole story, you're competing against tools that are getting better at that every quarter. If your pitch is "we understand your business well enough to know where your next customers are coming from and how to reach them efficiently," AI actually makes you more valuable, not less.
What this means for businesses evaluating their marketing
If you're a business owner or marketing leader, this is actually good news. AI gives you a lens for evaluating where your marketing spend is going. Ask your agency or team a simple question: what has changed in how you work since AI tools became widely available? If the answer is "nothing," that's a red flag. Not because they should be using every new tool, but because the landscape shifted and they either didn't notice or didn't adapt.
Ask what they're doing with the time that AI is saving them. Are they going deeper on strategy? Testing more aggressively? Bringing you insights faster? Or are they doing the same work in the same way and billing the same hours?
The best marketing partners right now are the ones using AI to amplify their strategic capabilities. They're still human-led, judgment-driven operations. But they're faster, more responsive, and more proactive than they were two years ago because they're using AI for what it's good at and focusing their human attention on what it can't do.
The bottom line
AI isn't going to replace your marketing team. But it is going to make it very clear whether your marketing team is providing strategy or just providing labor. For businesses that are paying for genuine strategic value, AI makes that partnership stronger. For businesses that are paying for execution dressed up as strategy, AI is going to surface that gap whether anyone wants it to or not.
The question worth asking isn't "should we be using AI in our marketing?" It's "does our current marketing operation provide strategic value that AI can't replicate?" If the answer is yes, invest more. If the answer is uncertain, it might be time for an honest evaluation of what you're paying for and what you're getting.
Jason Dellaripa is a media strategy leader with 20 years of experience across pharma, financial services, and regulated industries. Learn more or read about why marketing automation isn't a strategy.