Meta is rolling out an expanded set of generative AI tools for advertisers, after first announcing a set of AI features last October. Now, instead of only being able to create different backgrounds for a product image, advertisers can also request full image variations, which provide AI-inspired ideas for the overall photo, including riffs that update the subject of the photo or the product being advertised.
In one example, Meta shows how an existing ad creative showing a cup of coffee sitting outdoors next to coffee beans could be modified to feature the cup, from a different angle, in front of lush greenery and coffee beans, evoking images reminiscent of a cafe. farm.
It might not be a big deal if the image is just meant to encourage someone to visit a local coffee shop. But if it was the cup of coffee itself that was for sale, then the AI variants offered by Meta could be versions of the product that didn’t exist in real life.
This feature could be misused by advertisers who want to trick consumers into purchasing products that don’t actually exist.
Meta admits this is a possible use case, saying an advertiser could tailor the result generated with the upcoming text prompt feature with different colors of their product, from different angles and in different scenarios . Currently, the “different colors” option could be used to trick customers into thinking a product is different than it is in real life.
As Meta’s example demonstrates, the coffee cup itself could be transformed into different colors, or could be shown from different angles, where each cup has its own distinct swirl of frothy milk mixed with the hot drink.
However, Meta claims to have strong safeguards in place to prevent its system from generating inappropriate advertising content or low-quality images. This includes “pre-guardrails” to filter images that its generation AI models do not support and “post-guardrails” that filter generated text and image content that does not respond not up to its quality bar or that it considers inappropriate. Additionally, Meta said it tested the feature using its full llama and ad image generation model with internal and external experts to try to find unexpected ways to use it, then fixed any vulnerabilities. found.
Meta says this feature has already started rolling out, and in the coming months advertisers will also be able to provide text prompts to accommodate image variations.
Image credits: Meta
Additionally, Meta will now allow advertisers to add text overlays on their AI-generated images with a dozen of the most popular fonts to choose from.
Another feature, image expansion, also introduced in October 2023, will now be available for Reels in addition to Feed, on Facebook and Instagram. This option leverages AI to help advertisers adjust their image assets to fit different image formats, like reels and feed. The idea is that advertisers could spend less time reusing their creative assets for different surfaces. Meta says text overlay will also work with image expansion.
One advertiser, smartphone case maker Casetify, said using Meta’s GenAI Background Generation feature led to a 13% increase in return on its ad spend. The company had been testing this option with its Advantage+ shopping campaigns, where AI features were first available in the fall. The updated AI features will also be available through Ads Manager through Advantage+ creative, as before.
Image credits: Meta
Beyond images, Meta’s AI can be used to generate alternative versions of the ad headline, in addition to the main ad text, which was already supported by leveraging the original copy. Meta says it tests how well this copy can also resemble the brand’s voice and tone, using previous campaigns as reference material. Text generation capabilities will be moved to Mets’ next-generation LLM (large language model), Meta Llama 3.
Full Generative AI capabilities will be available globally to advertisers by the end of the year.
Apart from the AI updates, Meta also announced that it will expand its subscription service, Meta Verified for Business, to new markets, including Argentina, Mexico, Chile, Peru, France and Italy. The service began testing last year in Australia, New Zealand and Canada.
Now, Meta Verified will offer four different tiers to its subscription plan, all with the basic features of a verified badge, account support, and identity theft monitoring. Higher tiers will include new tools like profile enhancements, connection building tools, and more ways to access customer support.
Meta Verified will soon be extended to WhatsApp, the company also said.
techcrunch
