Meta's AI ad tools create chaos, shift liability to brands

Meta is requiring advertisers to use its AI ad-generation tools, a shift that is producing mangled images, garbled text and false product claims, while the company explicitly places the blame for these errors on the brands themselves.

The practical results have been damaging for corporate reputations. Advertisers report the system producing people with twisted limbs and nonsensical products. Outdoor retailer REI faced a consumer backlash over an Instagram ad showing a bicycle with two handlebars, an error REI said occurred because Meta had "auto-enrolled" it in the feature, producing an "inaccurate" image.

Other cases show the tool altering creative strategy without permission. Consultant Jessica Gleim noted Meta suggested turning a client’s pyjama dress into a shirt and trousers, and adding men to a campaign for a women’s networking group in Montana. Marketer Abigail Hogue saw her own Valentine’s Day campaign warped by the system, leading contacts to accuse her of posting "AI slop". She requested a refund that has yet to arrive weeks later.

Meta’s official stance shifts all risk to the advertiser. A company spokesperson told reporters that "AI can make mistakes and that it is the advertiser’s responsibility to review the AI outputs." For businesses, this creates a serious legal and commercial hazard. An automated tool that alters a product or rewrites copy leaves a brand financially liable for claims it never approved.

This risk is compounded by technical instability. Several agencies report a persistent bug that automatically toggles AI settings back on after they are manually disabled. One agency executive said a Meta representative offered to check her ad IDs by hand ahead of a major launch, an admission that the automated flaw remained unresolved.

The platform trap

Despite the chaos, advertisers remain locked in. Meta’s advertising business generated roughly $196 billion last year and reaches 3.5 billion people daily. The company states that "millions of advertisers are finding value" in its Advantage+ tools. Because Meta possesses the data and scale to deliver results, most brands calculate they cannot afford to leave, giving the platform the leverage to dictate terms.

The issue reflects a wider industry struggle with automated marketing. Google labels AI-generated ads and uses its Performance Max tool to automatically crop and rewrite creative, though brands report it has largely avoided Meta’s most severe failures. Even Reddit is attempting to manage AI marketing "slop" using additional AI. Meta has also struggled with AI accuracy in consumer products, recently pulling a feature from its Muse Image model after a backlash and watching its own fake-image detector fail when images were cropped.