Digital advertising underpins the business model of the internet—keeping content free and accessible while powering a multi-billion-dollar ecosystem. Ad tech makes this possible, matching marketers with audiences at an unprecedented scale. But as the industry adapts to demands for higher value generation, the technology that supports it must evolve. Efficiency, precision, and scalability are no longer advantages; they’re necessities.
AI is powering this evolution, transforming the ad tech ecosystem by delivering hyper-relevant experiences and advancing the goal of total market efficiency. As one of the largest commercial applications of AI, ad tech is uniquely positioned to harness its capabilities and fuel innovation.
Ad tech’s scale is a playground for AI
Digital advertising expenditure, larger than the cloud computing market, is set to exceed $700 billion annually and grow for the foreseeable future. Ad tech’s vast amount of data, significant budget, and direct applicability to consumer experiences underscore the reasons for the industry’s AI investment. Innovations have already led to improved market efficiency, better value, the ability to process larger data sets, and ultimately, better experiences for marketers, media owners, and consumers alike.
Three key characteristics fuel AI’s potential in ad tech:
1. Computing efficiency: Recent advances in AI and large language models (LLMs) have driven corresponding advances in computing, scale, and efficiency. Massive investments have led to significant price and performance improvements, outpacing Moore’s Law. Connectivity and infrastructure innovations make it possible to deploy tens of thousands of graphical processing units (GPUs) on a single workload. In turn, this allows AI to economically process and extract business insights from increasingly larger datasets.
2. Enormous datasets: At Index Exchange, we ingest two petabytes of data and process 550 billion transactions (and growing) daily. This data contains valuable signals that, when interpreted correctly, can be used to deliver more value to customers. However, traditional big data systems and streaming architectures struggle at this scale, often taking longer to generate insights than the data remains relevant. As demonstrated in modern LLMs, training and fine-tuning models enable us to perform an efficient lossy compression (a type of data compression that permanently removes some of the original data to reduce its size) on vast amounts of data. This allows us to extract insights from a significantly smaller model, facilitating faster query processing and more economical data systems.
3. Shrinking margins: Global digital advertising workloads will continue to grow for the foreseeable future. Media buyers expect more value out of their spend while media owners seek to further optimize yield—leading to shrinking margins. The previously mentioned improvements in computing efficiency will reduce costs, allowing ad tech to lower transaction expenses and deliver greater value across the ecosystem.
How can ad tech benefit from AI innovation?
AI unlocks deeper insights from data, enabling both the buy side and sell side to gain faster, previously unattainable signals. Historically, AI innovation in ad tech has centered on the buy side, optimizing addressability through a range of signals—including cookies, the identifier for advertisers (IDFA), hashed email addresses, and IP addresses—and improving contextual relevance by analyzing URLs, app environments, and other content metadata.
Evolving privacy regulations along with platform restrictions from the likes of Google, Apple, and others are limiting the use of certain audience signals, making addressability more challenging. At the same time, the surge in video content across the digital landscape is changing how the industry captures and shares contextual signals.
Up until now, the sell side has remained largely commoditized—but that dynamic is shifting. There’s an opportunity to incorporate more AI-driven intelligence on the sell side to further enhance buy-side optimization and deliver stronger outcomes amidst ongoing signal loss.
Stronger outcomes for media buyers
AI enables more precise targeting, helping marketers reach the right audiences faster, improving outcomes and reducing wasted ad spend. The next generation of AI will revolutionize demand-side targeting by:
- Enhancing content classification where metadata is sparse, for example in video
- Improving ad selection and placement with deeper contextual insights
- Ensuring accurate cohort targeting while navigating privacy-enhancing technologies
Improved yield for media owners
Beyond audience addressability, sell-side optimization must recognize broader, community-level trends that shift over time. Consider search behavior: 15% of Google searches each day have never been seen before, illustrating how consumer intent is constantly evolving.
AI can unlock more intelligent decision-making capabilities and improve inventory packaging and pricing, helping to increase CPMs and boost yield, especially in high-value formats like video. AI can drive more value for media owners by:
- Analyzing shifting content trends and providing media owners with feedback to enhance their inventory’s appeal and reducing the time needed for a feedback cycle
- Classifying inventory into relevant, easy-to-purchase packages for buyers, with AI keeping them updated as traffic patterns evolve to maintain scale and relevance in a brand-safe environment
Transforming untapped data into actionable insights
Ad tech is sitting on a wealth of untapped data—real-time traffic patterns, bidding insights, and auction outcomes—that often goes unused due to storage and computing constraints. AI offers the transformative power to harness this data, delivering predictive algorithms that refine targeting, optimize ad placements in real time, and uncover emerging consumer behaviors. By lowering costs and enhancing relevance, AI is driving a smarter, more efficient advertising ecosystem.
Harnessing AI’s potential creates a future where both the buy side and sell side can operate more effectively and with greater precision, all while preserving high-quality consumer experiences.
Our teams are continuously innovating to lead the industry toward total market efficiency, resulting in higher win rates and maximized yield, all at a lower cost. Learn more about our approach to exchange efficiency.
