At Cannes Lions this week, six companies - Alembic, AWS, Criteo, Higgsfield, KERV.ai and Taboola - are demonstrating how NVIDIA's AI infrastructure powers autonomous operations across advertising and marketing. The demos range from causal modeling that proves marketing ROI to creative production agents that handle full campaign lifecycles, signaling a shift from AI experimentation to enterprise-scale deployment in the industry.
Decision intelligence at enterprise scale
Causal AI platform AI for Marketing specialist Alembic helps enterprises answer a persistent question: which marketing initiatives actually drive growth, not just which ones correlate with it. Its models require AI infrastructure that processes enormous, fast-changing datasets without reducing them to simple correlations. Alembic will be the first causal AI company to use NVIDIA DGX Vera Rubin NVL72 systems and SuperPODs for enterprise-scale modeling, giving executives a single source of unbiased truth on what drove business outcomes and where capital is being wasted. The inference runs on private supercomputing infrastructure inside Equinix data centers where enterprise data already lives, and World Wide Technology extends this to secure and regulated environments.
Smarter bidding at auction speed
Amazon Web Services (AWS) is combining cloud infrastructure, foundation models, and NVIDIA GPU-accelerated computing into a reference implementation for adtech. The stack uses NVIDIA Triton Inference Server to deliver deep learning inference fast enough to fit within real-time auction windows, letting adtech companies move from rules-based bidding to AI-powered models for bid price optimization, audience activation, and deal scoring directly in the live pipeline.
Advertising company Criteo works with retailers to show the right product to the right shopper across one of the largest recommendation networks. Continuous retraining on billions of shopper timelines is critical, and speed translates directly to quality. Collaborating with NVIDIA, Criteo achieved a roughly 2x speedup in model training on NVIDIA Blackwell GPUs using the NVIDIA cuEmbed open library, freeing approximately 17,000 GPU hours a year. Taboola applies the same infrastructure logic to conversational AI, using NVIDIA GPUs to power DeeperDive, its AI answer engine, and extending that infrastructure to AI platforms and chatbots so they can generate revenue from advertising.
Agentic AI across the marketing workflow
AI agents are increasingly acting as digital coworkers across planning, execution, and optimization. The NVIDIA Agent Toolkit, including NemoClaw blueprints and the OpenShell secure runtime, provides safety guardrails, auditability, and role-based permissioning needed for enterprise deployment. Higgsfield AI's Supercomputer agents manage the full marketing automation lifecycle - from campaign ideation, planning, creative production to posting and autonomous optimization - in a single interface, orchestrating large language models alongside 35+ image, audio, and video models. AI Agents & Automation like these are integrated with NVIDIA Agent Toolkit software, giving the platform's enterprise trust layer. Marketing campaigns for nearly 400 Fortune 500 companies are created on the platform.
Contextual and content intelligence at scale
KERV's Moment Match Engine evaluates signals across every video frame to understand individual scenes, objects, and products, providing content recommendations based on ad creative. The company recently optimized its processing pipeline, achieving over 10x improvements in speed and efficiency when using the NVIDIA Nemotron 3 Nano Omni open model. On the MediaPerf benchmark for AI video understanding, that model delivered the highest throughput and lowest inference cost of any model evaluated, open or closed source. KERV's solution analyzes what each ad or media brief contains, who it resonates with, and which moment within content environments to target.
Why this matters for marketing professionals
For marketing leaders, the Cannes demonstrations show that AI is no longer just a creative assistant but an infrastructure play. Faster model training, automated bidding pipelines, and agentic workflows are becoming table stakes, not differentiators. The companies that treat AI deployment with the same rigor as any enterprise tech stack - prioritizing security, auditability, and measurable ROI - will gain a sustained advantage as the industry moves from experimentation to production-grade operations.
Your membership also unlocks:
