by Is Meta asking you to hand over more control of your Facebook ads than you're comfortable with? Wondering which of its new AI tools you can actually trust, and which ones still need a human in the loop?
In this article, you'll discover how to use Facebook's new AI-powered tools to change the way you track, build, and manage your ads, and where you should still maintain human control.
What Meta's Shift Toward AI-Guided Ads Means for You
With every passing month, Meta asks marketers to give up more control over their ads, across tracking, creative, campaign management, and reporting. The trade-off is the promise of better results. The catch is that it puts more responsibility on you to know what to trust, what to double-check, and what to keep your hands on.
Nick Theriot has watched this play out from the front lines as an agency owner who runs ads for e-commerce brands. His overall read is that advertising gets easier every day because the barrier to entry keeps dropping. Almost anyone can set up a campaign now, even without strong technical skills.
The direction is clear: Meta is moving marketers away from manual control and toward guided control, where you tell AI what you want and it does more of the clicking. The sections below cover where that shift helps, and where Nick still keeps a human in the loop.
#1: Let AI Handle Your Facebook Pixel Setup
The Facebook pixel is the tracking code that tells Meta who visits your site and what they do there. Meta is now making the pixel smarter with AI, so it can automatically connect and send data from your pages and products, including details like product names and availability, with far less manual setup. For many business owners, that removes the need to hire a developer just to get tracking working.
Nick sees this as exactly the kind of job AI is built for. He compares it to using AI to code an entire website in seconds, rather than hiring a developer and jumping through hoops. Pressing a button or two and letting AI plug the pixel in everywhere is a black-and-white task he's happy to hand over.
You still keep some control. You can turn the AI off, and you can tell it which categories of data to pull from your site and which to leave alone.
If you're not running Facebook ads, Nick says there's no need to rush a developer to set this up. But if you're even thinking about running ads, he recommends installing the pixel now so that it can start learning who actually buys from you and who doesn't. When you are ready to run ads, you’ll already have that data, so you'll spend less to optimize Meta's algorithm.
How much the pixel matters depends on your business.
For an e-commerce business selling physical products, he considers the pixel highly important. Most of Nick's ecommerce clients use Shopify, which offers one-click integration.
A media or publishing site is effectively its own version of Facebook, competing for the same attention, so the trade-offs there are different.
For lead-generation businesses with custom funnels, multi-step landing pages, and one-time upsells, setup gets more complex, and that's where the AI-powered pixel helps most.
Pro Tip: Tools like Google Tag Manager let you control exactly which pages fire the pixel.
Which AI Moves Actually Matter?
That's what every marketer is wrestling with now. New AI strategies, new tools, new takes every week — but no clarity. Most marketers and business owners are trying to figure out AI alone.The AI Business Society is your trusted guide. Get expert-led training you can put to work immediately. Plus a community of marketers sharing what's actually working.
I'M READY FOR REAL AI RESULTS#2: Connect AI Agents to Your Ad Account, Carefully
Meta is bringing more AI agents and connectors into the picture, letting marketers manage campaigns from inside third-party AI tools, including creating, editing, and analyzing them. Campaign setup is less about clicking every button yourself and more about telling AI what you want done.
You can already connect a tool like Manus to Meta, and Nick believes a Claude connector launched recently as well. Manus can work with Ads Manager to turn ad performance data into dashboards, slides, reports, and insights. On Instagram, it can ideate, generate, publish, and analyze content across posts, carousels, stories, and reels, with similar capabilities expected to reach Facebook.
Nick urges caution here. These connectors are still entry-level, and over the last two months, he's seen many people report that their ad account got banned right after connecting it to an AI. His read is that people are spamming the AI with a high volume of requests, Meta sees that flood of activity hitting the account, and it freezes things as a safety measure. As soon as he saw that pattern, he immediately disconnected his own ad account from both Claude and Manus. He'd connect again, but only carefully.
He's also clear about what he won't hand to AI: the human side of research and high-level ideation. Ask AI for target customers, and it tends to return generic segments. Nick still wants to understand why he should target one customer over another, or prefers to do his own research on new audiences a product can serve. Seeing real demand with your own eyes gives you more conviction and a better sense of how to open the conversation in your creative.
As these tools become more ubiquitous. Nick expects the traditional media buyer role to fade, while the marketing manager role will rise in its place. Marketers will coordinate an AI for the ad side and others for creative and landing pages, while keeping a human hand on performance. Looking two years out, he says the most valuable skills will be creating offers that scale, communicating your product's value, and grabbing attention while AI executes
Pro Tip: Treat working with AI like a normal conversation. Turn on your mic and tell it exactly what you want and what you dislike about each output, rather than laboring over a complex prompt for days.
#3: Use Meta's AI Business Assistant as a Smart Second Opinion
Meta is expanding its AI Business Assistant, which works like ChatGPT built into Ads Manager, surfacing recommendations and insights as you work.
Nick says 90 to 95% of that advice will be better than anything a beginner could piece together from YouTube videos and blog posts. For someone new to ads, it can cut years off the learning curve and level the playing field, without spending thousands on consulting calls or an agency.
The catch is that the assistant is trained on Meta's own rule book, so it tends to behave like the Meta reps who call advertisers, and Nick has heard plenty of bad advice from those reps over the years. For example, while he was optimizing for purchases and scaling an e-commerce brand, a rep pushed him to run a link-click campaign for more cold traffic. His team tested it and never saw the promised results.
He's most skeptical when the advice is to spend more money. A recommendation to increase your daily budget overnight because your cost per result looks good rarely holds up. Nick says you can't multiply your budget that fast and expect the same cost per result, and the outcome is usually worse. His rule of thumb is to be extra skeptical of “spend more” suggestions and to use your judgment otherwise.
Pro Tip: Check the AI's math, since not every model is equally reliable at it. As a practical pointer for auditing your numbers, Gemini tends to be weak at math when analyzing spreadsheet data, while Claude is unusually strong at it, likely because Claude is also built for coding, and math and code are both structured languages. When you're spending something like a million dollars a month at a 0.8 return on ad spend while acquiring customers on a twelve-month payback window, being off by a few dollars in your math can sink a company.
3 Days of World-Class Training—Zero Travel!
Couldn't make it to Social Media Marketing World and AI Business World this year? Get all of the great content at a fraction of the price with a Virtual ticket.
That’s full access to recordings of every keynote, workshop, and session—the ones people travel thousands of miles to see. Don't wait. Get your Virtual ticket and enjoy actionable content that you can watch anytime, anywhere.
#4: Produce AI-Assisted Creative to Stand Out
Creative is now the main driver of ad performance. Nick says the brands that have won over the last few years are the ones that put the customer first in their creative, because prospects never see your account structure. They only see what you put in front of them, and now Meta itself is pushing creative as the lever that matters most.
That's a shift from the 2018-to-2019 era, when success meant complex account structures with multiple campaigns, cost caps, and bid caps. By roughly 2020 to 2021, the brands testing more creative were the ones winning. Today, the structure is simple: one condensed campaign, a batch of creatives, and continuous testing to find winners and scale.
With that in mind, Nick pushes back on the advice to test 100 or 200 creatives a week. He's watched people do all that work, get no results, and feel like they were lied to, because they keep producing the same mediocre-looking ad on a mass scale. When his team tests less but with more intention to make something genuinely new and different, performance jumps. Viral organic content works for the same reason: it's new, original, and different.
He frames it simply: AI amplifies you. If you already have good ideas, AI helps you execute them faster and produce more of them. If your ideas are weak, AI just helps you produce more weak material faster. This is also why people with a strong creative background, such as photographers, video creators, and experienced copywriters, often get the best results from AI image and video tools. They know how to direct a model toward something original, rather than accepting the generic, recognizable look these tools produce straight out of the box.
Nick's team has dramatically cut the time from idea to finished ad.
Tools That Speed Up Meta Ads Creative Production
First, Nick's team has largely phased out writing ad copy from scratch. AI now writes about 90% of it, and the team does what he calls copy chiefing, tweaking the final 10% to sharpen it. However, in cases where they know exactly what they want to say, they still write it themselves.
AI Tools for Ad Images and Graphics
- ChatGPT image generation: The current version, ChatGPT Image 2.0, has become a standout in this space, ahead of Gemini's Nano Banana. For a recent Memorial Day sale, Nick's team plugged a client's website into ChatGPT, asked for a set of static ads highlighting the offer, and gave it one example to work from. It pushed out 5 to 10 static ads that they could upload directly to Meta.
- Higgsfield: Nick's team uses it to create AI UGC and graphic ads. It works as an aggregation tool that pulls together different creative capabilities on the back end.
AI Tools for Ad Creative Workflows and Automations
- Manus: It's easy to use and easy to build repeatable creative workflows in.
- Claude Cowork: This tool is good at chaining together a series of permitted actions.
Meta is also rolling out tools to create UGC-style videos, AI voiceovers, and translations, including realistic clips of a person talking about a product that can look like they came from a real customer, even though they're AI-generated.
Nick offers some useful insight here. Using an AI persona as a spokesperson to talk through a product, its ingredients, and its benefits is fair game. Using an AI persona to make fabricated claims, such as saying they took a supplement for thirty days and lost thirty pounds, is the kind of thing that gets advertisers sued.
Pro Tip: New York is set to become the first state to require AI disclosure, with a law taking effect June 9, 2026, that mandates a disclaimer whenever an ad on any platform, including Facebook, YouTube, Google, or even television, features an AI-generated person. Nick expects more states to follow and supports the change.
#5: Use AI Shopping Tools to Shorten the Path From Ad to Sale, Without Killing Conversions
Meta is also adding shopping tools to the middle of the funnel:
One-Click Checkout: Someone can click Buy Now and complete the purchase instantly with their payment details already saved.
Post-Click AI Shopping Features: Surface product details, reviews, pricing, and recommendations after a click, drawn from your site or from other metadata, and even a buyer profile.
Nick’s opinion of these unproven tools is measured. He remembers Meta introducing Facebook Shops a few years ago to little success. He's seen one brand make some sales from it, but no brand really takes off, pushing it.
His view is that it comes down to the product.
One-click checkout suits simple, low-ticket items people buy out of habit, like a plain t-shirt or cooking utensils, where the thought is just “I ran out, let me grab some.”
More considered purchases work better with a multi-step landing page and more product education. For those products, buyers want to leave and do their homework: Google the company, check Reddit, look for a refund guarantee, and see how the brand handled customers who had problems. That research is part of the sales process, and stripping it out can cost you the sale.
Pro Tip: On Shopify product pages, when Nick’s team changes the button from “Add to Cart” to “Buy Now,” they consistently see a lower conversion rate. There's something psychological about the add-to-cart step: it gives the buyer a two-to-three-second buffer to think before checkout, and because it's a familiar habit, it feels safe. Replace it with an immediate buy button, and the money is gone right away, disrupting that natural rhythm. Before you adopt one-click mechanics, know that this trap exists.
To manage all of this without betting the business on an unproven feature, Nick follows an 80/20 rule. He puts 80% of his time, energy, and budget into what's working and making money right now, and 20% into exploring new options like Meta's shopping tools. Some of those new bets will take off, and some are just hype that fades.
Nick Theriot is the founder of Theriot Solutions, an agency specializing in helping eCommerce businesses scale with ads. His course is Next Tier Ads, and his eBook is Instant Inbox Placement. You can find him on YouTube and Instagram.
Other Notes From This Episode
- Connect with Michael Stelzner @Stelzner on Instagram and @Mike_Stelzner on X.
- Connect with Jerry Potter on LinkedIn and YouTube.
- Watch this interview and other exclusive content from Social Media Examiner on YouTube.
Listen to the Podcast Now
This article is sourced from the Social Media Marketing Podcast, a top marketing podcast. Listen or subscribe below.
Where to subscribe: Apple Podcasts | Spotify | YouTube Music | YouTube | Amazon Music | RSS
✋🏽 If you enjoyed this episode of the Social Media Marketing podcast, please head over to Apple Podcasts, leave a rating, write a review, and subscribe.
Stay Up-to-Date: Get New Marketing Articles Delivered to You!
Don't miss out on upcoming social media marketing insights and strategies! Sign up to receive notifications when we publish new articles on Social Media Examiner. Our expertly crafted content will help you stay ahead of the curve and drive results for your business. Click the link below to sign up now and receive our annual report!
Curious About How to Use AI?
Our newest show, AI Explored, might be just what you're looking for. It's for marketers, creators, and entrepreneurs who want to understand how to use AI in their business.
It's hosted by Michael Stelzner and explores this exciting new frontier in easy-to-understand terms.
Pull up your favorite podcast app and search for AI Explored. Or click the button below for more information.
