Making Meta ads is about to get much faster. The risk is that a false claim, off-brand image or expired offer can multiply just as quickly.
Meta is building an advertising workflow that can diagnose creative, generate variations and optimize campaigns in one place. Its new Brand Memory feature is designed to learn an advertiser’s identity and tone from existing ads.
The direction is significant, but the new system is not broadly available yet. Meta said on June 23 that the end-to-end creative solution and Brand Memory are being tested, with wider expansion planned over the coming months. A separate integrated approval flow is also in testing, while the Meta Creator Marketing Hub is expected later in 2026.
Small advertisers have time to prepare. Start by creating a reliable approval process for the claims, images, offers and brand rules that Meta’s tools will use, before producing hundreds of AI variations.
Meta announced several products with different timelines
Meta presented the updates together at Cannes Lions 2026, but they do not form one finished product that every advertiser can use today.
Meta capability
Current status
What it is designed to do
Advantage+ creative tools
Available in eligible accounts; features vary by account and ad type
Generate text, expand images, create backgrounds, animate static images and add music
End-to-end creative solution and Brand Memory
In testing, with wider expansion planned over the coming months
Diagnose creative needs, develop variations and use existing ads to learn brand identity and tone
Integrated creative approval flow
In testing
Bring feedback and approval into the creative workflow
Meta Creator Marketing Hub
Planned for later in 2026
Combine Creator Marketplace and Partnership Ads Hub, with creator discovery and performance tools across Instagram and Facebook
Meta also announced expanded text-generation and translation capabilities. Those additions can increase the number of versions an advertiser produces, but availability can differ across accounts, placements and ad formats.
The Meta Business Agent Platform belongs in a separate category. Meta introduced it earlier in June for customer conversations, recommendations, booking and sales across WhatsApp, Messenger and Instagram. More than one million businesses already use Meta Business Agent, according to the company.
It is not the creative workspace or its approval system. Our analysis of Meta Business Agent covers the customer-service product and its data trade-offs.
Keeping these timelines and products separate prevents two costly assumptions: planning around a feature your account does not have and treating a customer-messaging system as an ad-production tool.
The performance figures do not prove the new creative tools work
Meta paired the Cannes announcement with a striking figure. Its analysis of more than one million campaigns during the week of April 13 to 20, 2026 found an average of $4.13 in advertiser revenue for every dollar spent on Meta ads, up 25% from 2022.
That is the average Meta reported from its large-scale analysis, not a forecast for an individual campaign. Revenue is also different from profit. A sale can look efficient in Ads Manager and still lose money after product cost, discounts, fulfilment, returns, agency fees and creative production.
Several other figures in Meta’s case concern ad ranking and measurement rather than AI-generated creative. Meta reported that its GEM recommendation model increased Facebook ad clicks by 3.5% and Instagram conversions by more than 1% during the fourth quarter of 2025.
It also credited Meta Lattice with a 12% gain in its internal ads-quality measure. The latest Incremental Attribution model produced 24% more incremental conversions than standard attribution, according to the company.
Those systems affect delivery, ranking or measurement across Meta’s ad platform. They do not establish that Brand Memory or the new end-to-end workspace will improve a particular advertiser’s profit. That result needs a controlled test using the advertiser’s own costs and customer data.
Creative influences delivery, but it does not replace targeting
The draft idea that creative “is targeting” goes too far. Meta’s own explanation of ad delivery says advertiser audience selections and the ad auction both influence who sees an ad. The auction weighs the advertiser’s bid, the estimated action rate and ad quality.
Creative can influence the last two factors. An image, message or offer may attract a specific type of person, giving Meta more response data and affecting where the system finds similar opportunities. Audience settings, exclusions, optimization goals, budget and placements still affect delivery.
The useful lesson is narrower: creative has become an input to both persuasion and distribution. Advertisers need distinct concepts that reveal different reasons to buy, not hundreds of near-duplicate headlines or background swaps.
A local contractor might test speed, workmanship and warranty as three separate ideas. A retailer might test price, durability and convenience. Those concepts give the system meaningful differences to evaluate while giving the business a readable explanation for the result.
AI can compress production time. It cannot decide which promise is true, which customer concern deserves attention or which sale produces enough margin to keep.
More variations make approval the bottleneck
When an advertiser can create dozens of assets quickly, review becomes harder to skip and more expensive to get wrong. One incorrect price can spread across formats.
A generated product image can show a feature the product does not have. A translated claim can become broader than the approved English or French wording.
Meta’s self-serve advertising terms place responsibility for ad content, targeting, placements and legal compliance on the advertiser. The company also describes some automated setup features as optional, with opt-in or opt-out controls where applicable. An AI label does not transfer that responsibility.
Meta says it provides AI information for ads created or significantly edited with its generative tools, with the label’s placement depending on the content and type of edit. Disclosure cannot confirm that a price, testimonial, product image or performance claim is accurate.
Canadian advertisers also remain subject to the Competition Act. The Competition Bureau says marketing to Canadians cannot be materially false or misleading, and that regulators consider an ad’s general impression as well as its literal wording. An AI-generated image, video or line of copy can contribute to that impression.
Before approving an AI-assisted ad, compare it with a fixed source of truth for current prices, product specifications, offer dates, warranty terms and approved claims. Confirm rights for logos, music, people and creator content. Send regulated, comparative or high-risk claims through the appropriate legal or compliance review.
Platform controls still need an account-level audit
Marketing Brew reported in April that some agencies found unwanted AI creative changes difficult to locate and disable. A Meta spokesperson told the publication that advertisers could opt out of ad-creative test features through Ad Account Settings and remained in control of whether the features were used.
Both points can be true across different accounts, tests and interfaces. A settings page may offer an opt-out while a busy team misses a newly introduced control or assumes that a choice made at the campaign level applies everywhere.
Do not base the plan on speculation that Meta will punish every advertiser who opts out. Meta’s ad terms do not guarantee reach or performance either way. Base the decision on a current account audit and a measured campaign result.
Check creative-enhancement settings when a campaign is built, again before launch and after any duplication or major edit. Preview every placement rather than only the primary feed version.
Record which automation was active, who approved it and when the account was checked. That record makes a later performance or compliance problem far easier to investigate.
Creator tools solve a different production problem
The Meta Creator Marketing Hub is meant to combine creator discovery, partnership-ad management and performance insight. Meta says the hub will bring its Creator Marketplace and Partnership Ads Hub together later in 2026, extend marketplace access to Facebook creators and build on the more than five million Instagram creators already in its marketplace.
The hub may reduce the administrative work involved in finding and using creator content. It does not remove the need to vet the creator, agree on usage rights, approve the claim and confirm disclosure requirements. Our guide to influencer marketing and the creator economy explains the relationship and campaign decisions that still sit with the business.
Treat creator content as another governed input. Document where the asset can run, how long the licence lasts, whether it can be altered with AI and who approves the final version.
Run one controlled AI creative test
Small businesses do not need agency-sized asset libraries to learn from Meta’s automation. One structured test can reveal whether the tool improves useful results without giving up control.
Choose the business result. Use gross profit, qualified leads, booked appointments or another outcome connected to revenue. Record the current cost and conversion rate before changing the creative process.
Freeze the source material. Create an approved file for prices, offers, product facts, brand rules, prohibited claims and visual rights. Give the AI only material that is current and safe to use.
Test meaningful concepts. Compare a small number of different customer motivations rather than a large batch of cosmetic variations. Keep the objective, audience, budget and test period as consistent as the account allows.
Review every generated asset. Check the image, copy, destination, call to action, translation and placement preview. Reject any version that invents a feature, weakens a qualification or misrepresents the product.
Limit the first exposure. Set a budget cap and inspect delivery early enough to stop a bad variation before it consumes the test budget.
Read the result outside Ads Manager. Compare platform-reported conversions with sales, margins, lead quality and customer records. Keep the winning concept only if the business result improves.
Build the approval system before Brand Memory arrives
Meta is reducing the production work required to make ad variations. Its latest announcement does not make brand judgement, factual review or financial discipline automatic.
That leaves advertisers with a practical job before the new creative workspace reaches more accounts: decide what the AI may use, what it may never claim, who approves the output and which business metric determines whether a variation survives.
Write those rules now. Then test Meta’s available AI creative features against one controlled campaign. Faster production is useful only when the business can trust what it is producing.
Affiliate disclosure: Some links in this post are affiliate links. See full disclosure in the page footer.
Not sure what to read next?
I can suggest related Tech Help Canada articles based on the topic you’re reading now.
We empower people to succeed through practical business information and essential services. If you’re looking for help with SEO, copywriting, or getting your online presence set up properly, you’re in the right place. If this piece helped, feel free to share it with someone who’d get value from it. Do you need help with something? Contact Us
Want a heads-up once a week whenever a new article drops?
