NVIDIA's GPU technologies are enabling autonomous AI operations across the advertising and marketing industry, with comp

The digital era brought speed to advertising and marketing; the AI era is bringing autonomous operations. For companies building next-generation technologies in these sectors, the question is no longer whether to adopt AI but whether their infrastructure can support it at the speed and scale required.

At Cannes Lions in France running June 22-26, industry leaders including Alembic, Amazon Web Services, Criteo, Higgsfield, KERV.ai and Taboola are showcasing how NVIDIA technologies unlock greater creativity and enable faster, autonomous operations at enterprise scale.

Alembic is a causal AI platform addressing a core enterprise challenge: proving what marketing initiatives actually drive growth rather than just reporting correlations. Modeling true causation across every channel, market and audience requires AI infrastructure processing enormous, fast-changing datasets. NVIDIA DGX Vera Rubin NVL72 systems enable Alembic to scale causal AI models to analyze more variables, run larger simulations and quantify true growth drivers across marketing investments. Alembic will be the first causal AI company using NVIDIA DGX Vera Rubin SuperPODs for enterprise-scale modeling, giving executives unbiased truth on what drove business outcomes and where capital is wasted. The inference runs on private supercomputing infrastructure inside Equinix data centers where enterprise data already lives, keeping AI workloads local. World Wide Technology extends this to secure and regulated environments, offering a complete enterprise AI stack for executives and data leaders accountable for capital decisions.

Amazon Web Services is bringing cloud infrastructure, foundation models and NVIDIA GPU-accelerated computing together as a cohesive stack for the adtech industry. AWS gives advertisers, demand-side platforms, supply-side platforms and independent software vendors a production-ready reference implementation to run AI-powered bidding directly inside auctions using NVIDIA Triton Inference Server, which delivers inference fast enough for real-time auction windows. Companies can move from rules-based decisioning to AI-powered models for bid price optimization, audience activation and deal scoring within the live auction pipeline.

Criteo helps retailers show the right product to the right shopper at the right moment across one of the largest recommendation networks in digital advertising. Keeping recommendations relevant requires continuously retraining AI on billions of shopper timelines, where speed directly translates to quality. Collaborating with NVIDIA, Criteo achieved roughly 2x speedup in model training on NVIDIA Blackwell GPUs using the NVIDIA cuEmbed open library, freeing roughly 17,000 GPU hours annually.

Taboola applies similar infrastructure logic to conversational AI, using NVIDIA GPUs to power DeeperDive, its AI answer engine, and extending infrastructure to AI platforms and chatbots to generate revenue from advertising.

In marketing and other industries, AI agents increasingly act as digital coworkers on long-running tasks across planning, execution and optimization. These agents are deployable for enterprises only when they include proper controls: safety guardrails, auditability and role-based permissioning. The NVIDIA Agent Toolkit includes NVIDIA NemoClaw blueprints and NVIDIA OpenShell secure runtime providing these controls.

Higgsfield AI is an AI video and image generator production platform offering Higgsfield Supercomputer agents managing the full marketing automation lifecycle from campaign ideation and planning through creative production, posting and autonomous optimization in a single interface. It orchestrates leading large language models alongside 35 plus image, audio and video models including Higgsfield's proprietary Soul and Soul 2.0 models built on NVIDIA Blackwell. NVIDIA Agent Toolkit software, including NVIDIA Nemotron open models, powers specialized subagents within Higgsfield Supercomputer running continuously in every campaign. NemoClaw and OpenShell are being integrated to provide the enterprise trust layer. The result is the full marketing lifecycle from ideation through posting, performance analysis and optimization available in a single interface. Marketing campaigns for nearly 400 Fortune 500 companies are created on the platform.

NVIDIA's multimodal stack provides vector search, data processing and video understanding capabilities making advanced content-level AI intelligence viable at production scale. KERV.ai's Moment Match Engine evaluates multitudes of signals across every video frame and media asset to understand individual scenes, objects and products, providing content recommendations based on ad creative to drive improved engagement. KERV recently optimized its processing pipeline achieving over 10x improvements in speed and efficiency using the NVIDIA Nemotron 3 Nano Omni open model. The solution analyzes what each ad or media brief contains, who it resonates with and which exact moment within content environments to target. On MediaPerf, an open benchmark for AI video understanding, Nemotron 3 Nano Omni delivered the highest throughput and lowest inference cost of any model evaluated, open or closed source, adopted by ecosystem partners including PYLER using NVIDIA DGX B200 systems.