AI Is Running Your Paid Media Now – Here’s What The Experts Actually Think About That

Paid media has always been a discipline that rewards whoever understands the system best. For two decades, that meant knowing the platforms better than your competitors – which keywords to target, which audiences to build, how to structure campaigns. That competitive edge is disappearing fast, as platforms have largely taken those decisions over.

Google’s Performance Max and Meta’s Advantage+ now automate bidding, audience targeting and creative testing across entire platform networks from a single campaign. AI determines who sees an ad, when and where, without relying on the third-party cookie infrastructure that underpinned targeting for years. According to analysis of 2026 ad platform developments, smart bidding systems are adjusting bids in real time based on behavioural signals at a speed and scale no human team can match. Meta has indicated it’s moving toward full campaign automation and ad creation by the end of 2026.

The debate is over; AI has undeniably transformed paid media. The tough part – and what founders and marketing leads are really asking – is what that means for strategy, creative, measurement and where budget should go.

What’s Actually Changed

The transformation runs deeper than new campaign types. Performance Max runs across Search, Display, YouTube, Gmail, Discover and Maps from a single campaign structure. Advantage+ handles audience discovery and creative delivery across Facebook and Instagram with minimal manual input. Both systems are built on the same principle: feed the algorithm a conversion goal and strong creative inputs, and the machine handles the rest.

There’s a real upside to this shift – what used to take weeks of creative testing now only takes days. Audience segmentation that required manual hypothesis and iteration now happens automatically. Budget allocation responds to real-time signals rather than weekly human review. For businesses with clear conversion data and strong first-party data, these systems can outperform what a human team would achieve manually.

The complication is what happens when those inputs are weak, and what happens when the platform’s optimisation goals diverge from the advertiser’s actual commercial objectives. AI optimises ruthlessly for whatever goal it’s given. If the goal is set wrong, or if the data feeding the system is poor quality, the machine will optimise efficiently toward the wrong outcome. The current debate boils down to one question: are platforms optimised for their own efficiency, or for the advertiser’s bottom line?

We asked the experts.

Our Experts

• Chester Scott, Chief Strategy Officer, Lunio
• San Nakra-Shah, Co-Founder and Managing Partner, ChilliMint Europe
• Thomas Harpointner, CEO, AIS Media
• André Picart, Managing Partner and Co-Founder, Tribera
• Matt Benton, CEO, Real Time Marketing
• Kamel Ben Yacoub, Founder and CEO, Getuplead
• Max von Weber, CEO, adnomaly
• Charlotte Sheridan, Director, The Small Biz Expert
• Sam Dadd, Senior Paid Media Executive, Arke Agency
• Ashley Fletcher, CMO and VP of People, Adthena
• Neil Baker, Head of Media Europe, Tug
• Georgia Doyle, Senior Paid Media Manager, TAL Agency

Chester Scott, Chief Strategy Officer, Lunio

“The genuine transformation in paid media is the sheer scale and speed at which AI-driven campaigns, like Google’s Performance Max and Meta’s Advantage+, can test and iterate. When fed accurate data, these smart bidding algorithms are incredibly powerful.

“However, the idea that these platforms are a ‘set-and-forget’ magic bullet is massively overhyped. The biggest blind spot founders and marketers have right now is trusting the ad networks to grade their own homework. AI models are ruthless optimisers; they do not understand the difference between a high-intent human and a sophisticated bot. If your campaigns are polluted with invalid traffic or zero-value engagements, the AI will learn from those fake clicks and optimise your budget to find more of them. This drives up your true Customer Acquisition Cost while giving the illusion of efficiency.

“Right now, operators must stop treating ad fraud as a cybersecurity problem and start treating it as a foundational data hygiene issue. You cannot control the black box of AI bidding, but you can control the data going into it. Leadership should invest in third-party waste intelligence to independently audit and filter their traffic. By blocking invalid engagements before they hit the algorithm, you force the AI to learn exclusively from genuine human interactions, ensuring your budget drives actual business outcomes.”

San Nakra-Shah, Co-Founder and Managing Partner, ChilliMint Europe

“Yes, AI is transforming media buying. Targeting, bidding, optimisation and audience selection are increasingly handled by Google, Meta and other platforms, usually more effectively than humans can manage themselves. For the last twenty years, marketers have built expertise around understanding channels, audiences and optimisation techniques. The best teams won because they understood the mechanics of digital advertising better than their competitors.

“AI is rapidly eroding that advantage because every advertiser has ready access to the same optimisation engines and algorithms. These capabilities are becoming less like key differentiators and more like common utilities. AI is commoditising media buying and fundamentally changing where competitive advantage sits in marketing.

“But here’s the irony. As marketing becomes more automated, the most valuable skills are becoming more human. Creativity, judgement, storytelling, brand building and customer understanding are increasing in importance, because they’re harder to automate.

“The value is rapidly shifting from targeting towards ideas, content and creative development. The smartest brands have already recognised this. They are taking the efficiencies gained from media and investing more in content, creative production and customer insight. Because when every competitor has access to the same AI, the real advantage comes from knowing something about your customers that the algorithm doesn’t. Media buying is becoming a utility and customer understanding is becoming the differentiator. And that’s the one thing AI still can’t commoditise.”

Thomas Harpointner, CEO, AIS Media

“AI is absolutely making paid media better, but not because brands can simply ‘set it and forget it.’

“The biggest gains are coming from creative speed. AI helps marketers bring more ideas to life, produce more ad variations faster, lower the cost of creative testing and identify winning messages sooner. That matters because increasing spend behind weak creative does not improve performance. It just wastes money faster.

“Platforms like Google Performance Max and Meta Advantage+ can work extremely well, but only when marketers give them the right inputs. That means clean conversion tracking, clear revenue goals, strong first-party data and enough creative direction for the system to learn from. Without those inputs, automation simply scales whatever is already broken.

“What is overhyped is the idea that AI replaces paid media expertise. It does not. It moves the expertise upstream. Marketers now need a sharper understanding of the brand’s objectives, voice, style, audience and unique selling points. AI does not automatically understand brand personality or business strategy. Skilled marketers have to train, guide, evaluate and refine the output.

“For founders in 2026, the smart move is not to blindly increase paid media budgets because the platforms are ‘AI-powered.’ The smart move is to fund disciplined testing. Use AI to develop creative faster, test more angles, find what is working and what is not, then put more budget behind proven winners. AI rewards disciplined advertisers. It punishes lazy ones.”

André Picart, Managing Partner and Co-Founder, Tribera

“AI has genuinely changed the mechanics of paid media for the better. Major platforms are now incredibly good at optimising budgets, finding audiences and testing creative variations at a speed no human team could realistically match. In that sense, the technology absolutely works.

“Where I think people get carried away is assuming that faster automatically means better. AI can generate and optimise ads quickly, but it doesn’t necessarily make them memorable. In a feed full of content, blending in is a problem, and a lot of AI-generated advertising risks looking and sounding exactly like everyone else’s.

“Another challenge is transparency. As more of the process becomes automated, it can be harder to understand where results are actually coming from. Marketers are often asked to trust the platform’s recommendations, without always seeing the full picture behind the decisions being made.

“With that in mind, the businesses getting the strongest results are the ones using AI as an execution tool, rather than a replacement for strategy. They’re letting the platforms handle the heavy lifting, while staying closely involved in measurement, creative direction and decision-making. They know what success looks like before they launch a campaign, and they’re willing to challenge what the dashboard is telling them when something doesn’t feel right.

“In 2026, the competitive advantage won’t come from having access to AI – everyone will have that. The advantage will come from brand strategy, distinctive creative and the quality of the thinking behind the campaigns.”

Matt Benton, CEO, Real Time Marketing

“AI has simultaneously enhanced and obfuscated paid advertising. Platforms like Google Performance Max and Meta Advantage+ have delivered on their promise of efficiency for conversion-driven campaigns. Smart bidding gets to real business outcomes faster than any human, and automated creative testing workflows have shortened timelines from weeks to days.

“Where they fell short was predictably in control. The platforms take more and more control of the targeting and placement decisions that used to be made by campaign managers. Budgets that don’t have clean conversion tracking get thrown at ineffective placements that are difficult to diagnose. AI is only as good as the goal you set it. When your goal is not buying customers, it will efficiently optimise for something else.

“Measure before you spend is your best paid media strategy for 2026. Google Local Services Ads consistently deliver strong ROI for contractors because the intent is clear and a pay-per-lead model decreases variance. Stack Performance Max on top of your conversion-ready campaigns to scale your successes, not replace them. Ultimately, AI can’t help you if your follow-up is slow, and speed-to-lead is what determines whether your paid traffic converts.”

Kamel Ben Yacoub, Founder and CEO, Getuplead

“The challenge right now is separating AI noise from AI signal. In my first-hand experience managing B2B SaaS budgets, AI is transformative on the operational side but largely overhyped for core acquisition metrics.

“What’s actually working: significant productivity gains in AI-assisted workflows – reviewing search term reports, running structured account audits and generating ad copy drafts as a starting point. Ad creation is also the best use case. But optimisation for paid search is where it gets more complicated. Google rotates headlines and descriptions in different combinations but doesn’t tell you which specific combinations are actually performing. You get asset-level ratings, but no impression or conversion data per combination. So it’s difficult to assume your AI-generated headlines are better than your human-written ones.

“What’s overhyped: for B2B SaaS clients working with hundreds of conversions per month, full automation in bidding or targeting doesn’t improve performance. We haven’t seen better performance with AI-native tools that promise to improve performance with AI-assisted management and optimisation. Human expertise remains essential.

“Founders must combine AI analysis with expert manual review. Prioritise fixing the conversion funnel: once ads are solid, the landing page is almost always where things break down. A faster iteration loop between ads, landing pages and sales data will drive commercial results far better than automated bidding alone.”

Max von Weber, CEO, adnomaly

“The conversation around AI in paid media tends to focus on the upside: better targeting, faster creative, smarter bidding. What we see daily, working across paid social, programmatic and paid search, is the other half of the story.

“The most underrated risk isn’t that AI doesn’t work; it’s that AI is changing settings, expanding audiences and swapping creative elements in the background, often without the operator noticing. We’ve seen approved ads suddenly play music the client never signed off on. We’ve seen ‘AI-optimised’ audience expansions reach segments the brand explicitly excluded. We’ve seen creative variants pushed live that nobody on the team can fully account for. The front-end has never been easier; the back-end has never been more closed off.

“That’s the part you have to be careful with. When the interface becomes simpler and the platforms tell you to ‘trust the system,’ it’s tempting to take that at face value when the rest of your day is already stressful enough. But the back-end tells a completely different story and you only notice when something has already gone live, gone wrong or gone over budget.”

Charlotte Sheridan, Director, The Small Biz Expert

“Firstly, what anyone running ads on platforms such as Meta or Google needs to bear in mind is that those platforms want to make money. They are less bothered if their advertisers do, so many automations are designed to either get more clicks or more views – but not necessarily more sales for your business.

“Nor do automations take into account specific audience requirements. If you are in a regulated industry, be very careful, as using things like Advantage+ or AI Max may mean that you’re allowing those platforms to write your ad copy for you, which means you may no longer be compliant with any regulations.

“However, if you run an ecommerce fashion brand, allowing the platforms to get creative or find audiences for you can work exceptionally well. Rather than using your own hypothesis, AI can look at huge data sets and decide where your budget is best placed, what type of creative is likely to convert and when.

“It’s also important to note how manipulative some of the language now is when pushing users towards automation. Meta ads will say ‘further limit your audience’ when you want to add specific targeting, while Google will tell you that you may be ‘missing out on clicks’ by not using broad match terms. Missing out on clicks is not an issue. Paying for irrelevant clicks is the issue for most advertisers, as they will burn through budget with nothing to show for it.”

Sam Dadd, Senior Paid Media Executive, Arke Agency

“From our perspective, AI is absolutely transforming paid media, but it is not a replacement for strong strategy, clear testing and human review.

“What is working well is the ability to scale faster. Tools like Performance Max and Meta Advantage+ can be really effective when they sit within a wider marketing mix, particularly where there is strong conversion data, clear audience signals and a full-funnel strategy behind them. Smart bidding, automated placements and AI-led creative testing all have their place, but only when the foundations are already there.

“What feels overhyped is the idea that AI can simply be switched on and left to perform. It is only as powerful as the person behind it. Without proper brand safety checks, creative review, first-party data and clear measurement, AI can very quickly optimise towards the wrong outcomes.

“Right now, founders and marketers should not jump on the bandwagon for the sake of it. They should understand their audiences, test standard ad formats, compare manual and automated bidding, and build a strong A/B testing framework first. Once those foundations are in place, AI can support scalability and efficiency. The main takeaway is simple: use AI, but do not hand over control.”

Ashley Fletcher, CMO and VP of People, Adthena

“AI is reshaping paid media in 2026, with the biggest shift happening inside AI-generated search environments. Ads now appear directly in Google’s AI Overviews and ChatGPT conversations, marking a fundamental change in how search is monetised. Adthena was the first to detect these placements, spotting early Google AIO ads across 25,000 SERPs at just 0.052% frequency – a small signal of a major structural shift. That volume has since grown, with competition in AI ad auctions rising 35% year on year.

“What’s working today is AI-native optimisation combined with real competitive visibility across both Google and ChatGPT. Performance Max, Advantage+ and now Google’s emerging AI Mode roadmap are powerful, but they’re black boxes – and this automation is often mistaken for strategy.

“What’s overhyped is the idea that AI will ‘do it all for you.’ If anything, AI has made paid media more complex. Founders should prioritise cross-channel visibility, especially across Google and ChatGPT, where behaviour and ad delivery are already diverging. Success depends on rapid creative and bidding iteration grounded in real market data rather than platform-reported metrics, which increasingly obscure what’s actually happening in the auction.”

Neil Baker, Head of Media Europe, Tug

“AI is nothing new to performance marketing – since 2016 we have been using automated bidding strategies. However, the space is still evolving. With search we have intent-based strategies which means first-party data integration is more important than ever. With paid social, algorithm changes such as Andromeda show that creative is now the new targeting, and creative variation is incredibly important.

“Whilst creative variation is now more important than ever, that doesn’t mean making lots of variations of the same ad. It means developing creative themes that touch on the different values of your customer and allowing the algorithm to find the people who would relate. Don’t develop AI creatives just for the sake of it – use them with purpose.

“Another watch-out is people’s over-reliance on LLMs for strategy. The recommendations can often be dated and misleading and still need an expert to validate. The real value from AI comes in speed – if you can accelerate creative development or streamline workflows – and in measurement: MMM, incrementality modelling and causal modelling are where some of the most useful developments are happening.”

Georgia Doyle, Senior Paid Media Manager, TAL Agency

“AI, like anything in digital marketing, can be very useful for taking care of repetitive tasks and freeing up time for real creativity, so long as it doesn’t seep into creative and ad copy. As consumers become more savvy to how AI looks and feels, it’s crucial that paid media specialists understand its limitations and don’t let it take over creative aspects that require humans throughout the full process.

“AI helps with optimisation, audience expansion and bid management. Meta and Google can process far more data signals much quicker than any human could in real time, and features that improve efficiency and performance should absolutely be embraced. But the industry has become far too quick to assume that every AI feature automatically adds value.

“Paid media specialists should be very selective about what’s enabled and disabled across ad platforms. No one understands a brand as much as a human marketer, and marketing requires much context and nuance to ensure it’s landing well with an audience. One of the biggest issues I’m seeing is that AI-generated advertising is becoming increasingly recognisable. Consumers and marketers are becoming much more savvy to it, and AI-generated content is seen as inauthentic and can immediately erode trust.

“Many ads are starting to look and sound the same, with similar visuals, messaging and creative styles, which is off-putting to audiences and risks reputational damage. Differentiation is crucial, as consumers connect with brands when they feel authentic as opposed to sounding like every other advertiser using the same AI prompts. The strongest paid media strategies in 2026 are using AI as a tool for optimisation and efficiency, whilst keeping humans firmly in control of strategy, creativity and brand voice.”

For any questions, comments or features, please contact us directly.