Google Cloud announced a $750 million fund at its Cloud Next conference to accelerate agentic AI deployment across its 120,000-member partner ecosystem, as a string of major infrastructure deals and new AI tools underscored the intensifying race among technology giants to dominate enterprise AI.
The fund will support global consulting firms, systems integrators, and software partners through AI prototyping, agent deployment, and upskilling programmes. Google will embed forward-deployed engineers alongside firms including Accenture, Capgemini, Deloitte, and TCS, while partners such as Accenture, Bain, BCG, and McKinsey will receive early access to Gemini models. Kevin Ichhpurani, president of Google Cloud’s global partner ecosystem, described partners as increasingly critical channels for distributing AI technologies at enterprise scale.
On the infrastructure side, Mira Murati’s secretive startup Thinking Machines Lab — founded in February 2025 after Murati departed as OpenAI’s chief technologist — signed a multibillion-dollar agreement with Google Cloud, gaining early access to systems powered by Nvidia’s GB300 chips, which Google says deliver twice the training and serving speed of previous-generation GPUs. The deal, reported exclusively by TechCrunch, supports the reinforcement learning workloads underpinning Thinking Machines’ Tinker product, a tool that automates the creation of custom frontier AI models.
Google also unveiled generative AI features for its mapping and geospatial platforms, including Maps Imagery Grounding — which lets enterprise users generate realistic Street View scenes from text prompts via the Gemini Enterprise Agent Platform — and Aerial and Satellite Insights, which applies AI to satellite imagery stored in BigQuery, compressing weeks of analytical work into minutes.
