AI creative optimization is becoming the new operating system
AI creative optimization is the use of artificial intelligence systems embedded directly into advertising workflows to generate, test, and refine creative assets using live campaign data, so agencies can move from manual iteration to continuous, automated improvement across formats, channels, and audiences at scale.
The most important shift in agency work right now is not that AI can make ads; it is that AI is being wired into the tools agencies already live in. Meta’s new end‑to‑end creative solution sits on top of campaign data, generating and testing creative from a dedicated workspace while claiming that every dollar spent on its platforms returns $4.13 in revenue, a 25% increase since 2022. Instead of hopping between dashboards, teams can review performance, spin up new assets, and run experiments in one environment. This is less about novelty and more about speed, consistency, and the ability to industrialize creative decision-making without flooding teams with extra processes.
Meta and WPP: AI moves inside the agency brain
Meta’s integration into WPP Open is the clearest sign that AI advertising tools are evolving into workflow infrastructure, not side projects. WPP will be the first advertising company to test Meta’s AI-powered creative strategy and optimisation solution, directly within its own marketing platform. That matters because creative teams can now analyse performance, generate concepts, and test new campaign ideas using AI without ever leaving their native environment.
This is agency workflow automation in practice: the tool is being built to fit existing processes, so teams can create and optimise advertising without adding new steps or standalone tools. Meta’s “brand memory” capability then learns each brand’s identity and tone from past campaigns and applies those rules across AI-generated assets. The result is an always-on strategist that remembers the brand book and watches the numbers. For clients like Unilever, that means the promise of moving faster and “create smarter” work while staying on-brand, with a broader rollout expected in the coming months.
DoubleVerify: fixing the media quality trade-off with AI
Creative optimisation is pointless if media quality is broken. That is where AI advertising tools from verification players are quietly transforming the pipes. DoubleVerify has expanded its DV Authentic AdVantage solution to Meta and TikTok, extending AI-powered media optimisation across two of the largest social ecosystems. Instead of separate tools for safety, performance, and reporting, DV integrates pre-bid protection, AI-led optimisation, and independent measurement into one central layer.
The claim is bold: advertisers no longer need to choose between cost efficiency and media quality. DV says early TikTok deployments delivered a 98% improvement in unique reach, a 50% increase in efficiency, and a 59% reduction in brand suitability incidents. Those numbers turn AI creative optimisation into something measurable, not mystical. With DV’s platform combining verification, Scibids AI optimisation, and outcome measurement in a single operating fabric, agencies can plug AI ad generation and testing into media buying that is continuously tuned to reach, CPM, CPA, and attribution signals at scale.
AnyMind: hybrid human–AI production for social commerce
On the front lines of social commerce, the question is not whether AI will replace creators, but how to combine them. AnyMind Group’s AnyAI Video is a direct response: a tool designed to help brands produce AI-generated video content at scale alongside existing creator-driven campaigns. Built on its AnyAI data platform, it can output explainers, comparisons, reviews, tutorials, and FAQ content using platform insights and market intelligence to guide production.
Crucially, AnyAI Video is positioned as a complement, not a substitute. Creator content keeps doing the heavy lifting on awareness and trust, while AI-generated content focuses on product education and conversion. For BONCEPT, combining creator content, AI-generated video, and livestreaming led to about 50 content assets a month on TikTok, with AI videos contributing close to 10% of e-commerce GMV for the period. This is AI creative optimisation applied to volume and consistency: brands fill content gaps without burning out creators, while AnyMind’s wider stack—AnyTag, AnyLive, and AnyAI Insights—connects production to campaign data and even offline sales, supported by a new AI R&D hub in Hangzhou.
From tools to tactics: what agencies must change now
Across these moves, one pattern is clear: AI ad generation is shifting from optional add-on to default mode of operation. Meta is weaving AI into creative development, approvals, and multilingual copy, including expanded text-generation that can produce multiple messaging variations directly inside image ads based on past performance and business data. Its broader strategy is to embed AI across advertising, creative production, and customer communications, with features under active testing and wider rollout planned in the coming months.
For ordinary marketers, the impact is practical, not abstract. Agencies can now ideate, test, and analyse campaigns inside the tools they already use, instead of exporting spreadsheets and mockups. Media quality and optimisation are increasingly handled by AI systems in the background, reducing brand-safety risk while improving reach and efficiency. And social commerce teams can scale educational and conversion-focused content alongside creator output without sacrificing authenticity. The next competitive gap will not be who has access to AI, but who can redesign processes, roles, and measurement to let these systems run at full speed—without losing the human judgment that still decides what a brand should say.
