Mark Ritson recently told a room full of marketers to stop handing campaign decisions to AI agents they don’t understand. Harsh? Maybe. Wrong? Not really. The agentic AI warning lands at a moment when brands are automating budget shifts, creative swaps, and bidding decisions faster than their governance teams can review them. So what do you actually do on Monday morning?
Why Ritson’s Warning Landed So Hard
Ritson isn’t anti-AI. He’s anti-blind-trust. His argument, delivered at a marketing conference and amplified across LinkedIn within hours, boiled down to one uncomfortable point: marketers are deploying autonomous agents to make live spending and targeting decisions without understanding the failure modes. Not the capabilities. The failure modes.
That distinction matters. Every vendor pitch deck shows the agent succeeding. Nobody shows the demo where it reallocates 40% of budget to a channel because it misread a seasonality signal. Marketers have seen this movie before with programmatic, with automated bidding, with “smart” campaign types that quietly burned budget on brand terms nobody sanctioned.
The risk isn’t that agentic AI fails occasionally. It’s that it fails silently, at scale, inside systems nobody is watching in real time.
Marketing leaders now face a genuine tension. Slow down too much and competitors using agentic tools well will out-execute you. Slow down too little and you’re one bad prompt away from a six-figure mistake nobody catches until the invoice arrives.
What “Slow Down” Actually Means in Practice
Ritson’s advice isn’t “ban autonomous campaigns.” It’s “stop treating them like set-and-forget tools.” There’s a difference between pausing adoption and adding checkpoints. Most brands need the latter, not the former.
Here’s the practical translation: agentic AI needs the same governance rigor you’d apply to a junior media buyer with signing authority. You wouldn’t let a new hire reallocate six figures of spend without approval thresholds. Why would you let an algorithm do it?
- Define decision boundaries. What can the agent change autonomously (bid adjustments under 10%, creative rotation within an approved set) versus what requires human sign-off (budget shifts over a threshold, new audience segments, pausing live campaigns)?
- Build a kill switch that actually works. Test it. Most teams have a theoretical off switch nobody has pressed under real conditions.
- Log every autonomous decision. Not just outcomes, but the reasoning trail. If the agent shifted spend, you need to know why, not just that it happened.
- Set a review cadence separate from performance reporting. Performance reviews ask “did it work?” Governance reviews ask “did it behave as intended?” Those are different questions and need different meetings.
A Response Framework You Can Run This Quarter
Marketing ops teams don’t need another think piece. They need a checklist. Here’s a four-stage response plan that respects Ritson’s warning without freezing your roadmap.
Stage One: Audit What’s Already Autonomous
Most CMOs underestimate how much decision-making has already been handed to algorithms. Smart bidding, dynamic creative optimization, automated audience expansion, even CRM lead scoring: these all involve some degree of agentic behavior, even if nobody labeled them that way. Similar blind spots show up in CRM AI agent deployments, where teams assume automation is working as advertised without ever stress-testing the underlying logic.
Pull a full inventory. Every platform, every automated rule, every “smart” feature you toggled on because a rep recommended it. You cannot govern what you haven’t catalogued.
Stage Two: Classify by Blast Radius
Not all agentic decisions carry equal risk. A tool that auto-generates ten creative variants for A/B testing is low stakes. A tool that shifts your entire paid social budget across platforms based on real-time performance is high stakes. Rank each autonomous system by potential financial and reputational blast radius, then apply governance proportionally.
This mirrors the approval bottleneck problem many teams already face with creative production, detailed in this breakdown of unused creative: too much friction kills output, too little invites disaster. The goal is proportional control, not blanket caution.
Stage Three: Build the Governance Checklist
Brands running AI-driven social publishing have already had to answer these questions, and the frameworks transfer directly to broader agentic campaign tools. A solid governance checklist for AI posting agents covers approval workflows, brand voice guardrails, and escalation paths. Apply the same logic to media buying agents: define who approves what, how often audits happen, and what triggers a human review versus a fully automated pass-through.
Compliance considerations matter here too. If your agentic tools touch paid creative, you’re already operating in a space regulators are watching closely. The FTC’s guidance on advertising practices increasingly applies to algorithmic decision-making, not just human-authored claims. If your AI agent writes ad copy or selects claims autonomously, someone on your team needs to own the compliance sign-off, full stop.
Stage Four: Measure the Cost of Slowing Down
This is the part most governance frameworks skip. Adding checkpoints has a cost: slower campaign launches, more meetings, potentially missed opportunities. Quantify it. If a human review adds 48 hours to campaign launch but prevents even one five-figure misallocation per quarter, the math usually favors caution. Run the numbers for your own program rather than assuming either extreme is correct.
Where Marketers Are Getting This Wrong Already
Two failure patterns keep showing up. First: teams adopt agentic tools for efficiency gains, then never revisit the governance settings once the initial rollout is done. The agent that was reviewed carefully in month one is running unsupervised by month six because nobody scheduled a recheck.
Second, and more common: teams confuse “the vendor says it’s safe” with “we’ve validated it’s safe.” Every AI platform vendor, from ad tech to CRM providers, has an incentive to describe autonomous features as more reliable than they are. That’s not malice. It’s sales. But it means the burden of validation sits with you, not them.
According to eMarketer’s ongoing coverage of AI adoption in marketing, spend on AI-driven campaign tools continues to climb even as trust in fully autonomous decision-making lags behind adoption rates. That gap between usage and confidence is exactly the space Ritson is pointing at. Marketers are buying the tools faster than they’re building the guardrails.
There’s a parallel here with attribution work. Just as brands rushed to adopt blended attribution models before fully understanding what the blended output actually represented, many are now running agentic campaigns before understanding what “autonomous” really means inside their specific stack.
What This Means for Vendor Conversations
If you’re evaluating a new agentic AI platform, the sales conversation needs to change. Stop asking “what can it do?” Start asking “what happens when it’s wrong, and how fast will we know?”
Specific questions worth asking every vendor:
- What’s the maximum single-decision budget impact before a human is notified?
- Can we export a full decision log, not just a performance summary?
- Is there a documented history of the agent making a decision outside expected parameters, and how was it caught?
- Who at your company is liable if the agent’s decision causes financial or reputational harm?
Vendors who can’t answer these clearly aren’t ready for enterprise deployment, regardless of how polished the demo looks. This same scrutiny is why vendor scorecards have become standard practice in adjacent categories, as outlined in this AEO vendor evaluation framework. Agentic campaign tools deserve the same discipline, arguably more, since the financial stakes are usually higher.
The Middle Path: Supervised Autonomy
The realistic answer isn’t full autonomy or full manual control. It’s supervised autonomy: agents that execute within tightly defined boundaries, with mandatory human review at defined intervals and hard stops on high-impact decisions. Think of it less like replacing a media buyer and more like hiring one with a probationary period that never fully ends.
This isn’t a step backward. It’s how every mature automation category eventually settles. Marketing automation platforms went through this in the previous decade. Programmatic buying went through it. Agentic AI in marketing is simply the current frontier working through the same maturity curve, just faster and with higher stakes because the decisions now touch live budget in real time.
Brands that get this right in the next twelve months won’t be the ones who adopted agentic AI fastest. They’ll be the ones who built governance fast enough to keep pace with adoption. That’s the actual competitive advantage hiding inside Ritson’s warning.
Next Step
Don’t wait for a Ritson-style keynote to force the conversation internally. Run the four-stage audit this month, starting with a full inventory of every autonomous decision already live in your stack, and set a governance review date before you approve a single new agentic tool.
FAQs
What did Ritson actually say about agentic AI?
Mark Ritson warned marketers against deploying autonomous AI agents for campaign decisions without understanding their failure modes, arguing that speed of adoption has outpaced governance and oversight capability across the industry.
Does this mean brands should stop using agentic AI tools?
No. The recommended response is proportional governance, not abandonment: classify autonomous tools by risk level, set approval thresholds, and build audit trails rather than pulling back from automation entirely.
What’s the biggest risk with autonomous campaign tools?
Silent failure at scale. Agentic systems can make flawed decisions repeatedly across a live budget before anyone notices, unlike a human error that’s typically caught faster and contained to a single instance.
How do I know if my current tools are already “agentic”?
Audit any feature that makes real-time decisions without explicit human approval, including smart bidding, dynamic creative optimization, automated audience expansion, and AI-driven lead scoring. Many teams are already using agentic features without labeling them as such.
What questions should marketers ask AI vendors before adoption?
Ask about maximum single-decision impact thresholds, availability of full decision logs, documented history of out-of-parameter behavior, and clear liability terms if the agent causes financial or reputational harm.
Top Influencer Marketing Agencies
The leading agencies shaping influencer marketing in 2026
Agencies ranked by campaign performance, client diversity, platform expertise, proven ROI, industry recognition, and client satisfaction. Assessed through verified case studies, reviews, and industry consultations.
Moburst
-
2
The Shelf
Boutique Beauty & Lifestyle Influencer AgencyA data-driven boutique agency specializing exclusively in beauty, wellness, and lifestyle influencer campaigns on Instagram and TikTok. Best for brands already focused on the beauty/personal care space that need curated, aesthetic-driven content.Clients: Pepsi, The Honest Company, Hims, Elf Cosmetics, Pure LeafVisit The Shelf → -
3
Audiencly
Niche Gaming & Esports Influencer AgencyA specialized agency focused exclusively on gaming and esports creators on YouTube, Twitch, and TikTok. Ideal if your campaign is 100% gaming-focused — from game launches to hardware and esports events.Clients: Epic Games, NordVPN, Ubisoft, Wargaming, Tencent GamesVisit Audiencly → -
4
Viral Nation
Global Influencer Marketing & Talent AgencyA dual talent management and marketing agency with proprietary brand safety tools and a global creator network spanning nano-influencers to celebrities across all major platforms.Clients: Meta, Activision Blizzard, Energizer, Aston Martin, WalmartVisit Viral Nation → -
5
The Influencer Marketing Factory
TikTok, Instagram & YouTube CampaignsA full-service agency with strong TikTok expertise, offering end-to-end campaign management from influencer discovery through performance reporting with a focus on platform-native content.Clients: Google, Snapchat, Universal Music, Bumble, YelpVisit TIMF → -
6
NeoReach
Enterprise Analytics & Influencer CampaignsAn enterprise-focused agency combining managed campaigns with a powerful self-service data platform for influencer search, audience analytics, and attribution modeling.Clients: Amazon, Airbnb, Netflix, Honda, The New York TimesVisit NeoReach → -
7
Ubiquitous
Creator-First Marketing PlatformA tech-driven platform combining self-service tools with managed campaign options, emphasizing speed and scalability for brands managing multiple influencer relationships.Clients: Lyft, Disney, Target, American Eagle, NetflixVisit Ubiquitous → -
8
Obviously
Scalable Enterprise Influencer CampaignsA tech-enabled agency built for high-volume campaigns, coordinating hundreds of creators simultaneously with end-to-end logistics, content rights management, and product seeding.Clients: Google, Ulta Beauty, Converse, AmazonVisit Obviously →