Oracle’s AI Playbook: Co-CEOs & Explosive OCI Growth

Oracle just named two co-CEOs and doubled down on an AI-first future. Jeff Lazarto explains how OCI’s gigawatt-scale GPU superclusters, Oracle’s vectorized “AI database,” and application-layer AI agents could make Oracle the standard for AI training and—soon—enterprise inferencing. He also highlights what this means for customers evaluating Oracle today: timing, deal leverage, and a practical path from on-prem to cloud/Fusion.  

Key Points:

• Why big AI players are choosing OCI for training; cost/performance narrative.  
• Oracle’s bet that inferencing (AI agents doing work) will dwarf training.  
• The “AI database” and data privacy posture across models.  
• Leadership update: co-CEO model aligning apps vs. infra.  
• Customer takeaways: leverage, migration paths, and contract strategy. 

Episode Chapters

00:00:23 AI and OCI are driving Oracle’s momentum - Oracle’s earnings buzz is fueled by AI demand and OCI’s gigawatt-scale GPU superclusters selected by top AI players.

00:01:11 Why big tech is choosing OCI - Oracle claims faster/cheaper model training; efficiency and cost are key factors behind marquee customer selection.

00:02:04 Training now, inferencing next (the much bigger market) - Ellison frames inferencing—AI agents embedded in business processes—as the wave that will dwarf training.

00:02:46 From “college” to work: enterprise-tuned AI agents - Publicly trained models get fine-tuned on company data so AI agents can actually do tasks for the business.

00:03:05 Oracle’s AI Database and vector search - Oracle pitches a vectorized, privacy-preserving database that works across ~25+ models while keeping enterprise data secure.

00:03:41 Endorsement loop that benefits Oracle - If leading AI builders rely on OCI, enterprises may follow suit for hosting their own AI workloads.

00:04:24 OCI growth projections that shocked Wall Street - Oracle reiterated aggressive OCI revenue targets through FY30, a key driver of the stock’s surge.

00:05:12 Oracle’s stack: training → inferencing → data → apps - Strategy spans AI training/inferencing, the AI database layer, and SaaS built on AI application generators.

00:05:50 App generators = networks of AI agents - Oracle describes SaaS evolving from hand-coded software to AI agents linked by workflows.

00:07:07 Multicloud and “Cloud@Customer” options - Run Oracle in public regions, behind your firewall, or access Oracle Database from Azure/AWS/GCP via reseller arrangements.

00:07:28 Leadership shift: two new co-CEOs - Clay Magouyrk (OCI/engineering) and Mike Sicilia (industries/apps) step in; tech-first leaders for infra and apps.

00:08:05 Clay McGuirk’s remit (Gen2 OCI & AI DCs) - Architect behind high-performance OCI powering gigawatt-scale AI training is elevated.

00:08:30 Mike Sicilia’s remit (vertical apps & AI) - Industry SaaS leader focuses on applying AI within Oracle’s application portfolio.

00:09:12 Why tech visionaries at the top matters - Oracle follows the big-tech pattern: product-minded leaders setting direction vs. sales-led stewardship.

00:10:01 Co-CEO model returns; wider exec moves - Safra Catz shifts to executive vice chair; additional sales/finance leadership changes are in motion.

00:11:30 Two-engine company, two tech CEOs - Structure mirrors Oracle’s dual identity: infrastructure and applications led by AI-savvy chiefs.

00:11:57 Customer takeaway: leverage the moment - High expectations on Oracle create room to negotiate strong terms and strategic commitments.

00:12:44 Installed base path to AI value - Benefits of AI live in the cloud: Fusion for apps and OCI for tech; BYOL eases the move.

00:13:24 Expect harder Oracle push to cloud - Stronger GTM motions will highlight concrete AI/business value to drive migrations.

00:14:09 Start with your Oracle strategy, then engage - Define the enterprise roadmap first; invite Oracle to align capabilities and structure the right deal.