Contrary to fears of financial recklessness, Oracle's massive AI infrastructure spend is immediately profitable. Gross margins improved to 32%, beating guidance, demonstrating that strong customer demand and pricing power are covering the costs of their GPU buildout as soon as it comes online, avoiding dangerous financial engineering.
Larry Ellison argues that AI won't kill SaaS incumbents because Oracle is aggressively adopting AI coding tools internally. This allows smaller engineering teams to build new products and embed AI agents into existing suites more quickly, effectively neutralizing the speed advantage of new AI-native startups.
While user growth for apps like ChatGPT is slowing, per-user token consumption is skyrocketing as models shift from simple queries to complex reasoning and AI agents. This creates a hidden, exponential growth in compute demand, validating Oracle's massive infrastructure investment even as front-end adoption matures.
Larry Ellison's focus on Oracle's AI and his son David's acquisition of Paramount's IP is not a hedge. It's a unified bet that generative AI will enhance the value of existing intellectual property rather than make it obsolete, creating a future where both algorithmically generated and human-created content appreciate together.
The most damaging part of the Jones Act is its mandate for US-built ships, which are 5x more expensive than foreign ones. This stifles demand and kills the shipping industry. The proposed solution is to repeal this clause to drastically increase the number of US-flagged ships, thereby creating a robust domestic market for maintenance and, eventually, new shipbuilding.
A 1997 Vanity Fair profile depicts Larry Ellison's ambition as all-consuming, from a "burning ambition" to beat Bill Gates to heckling Michael Jordan. This historical context of extreme competitiveness and risk-taking is key to understanding why he is betting Oracle's future on an aggressive, all-in AI infrastructure buildout today.