IBM’s Hybrid AI Strategy: Renting GPUs from Cloud Giants While Boosting On-Prem Solutions

IBM’s AI-Driven Financial Turnaround

IBM has revealed impressive third-quarter 2025 results, showcasing how artificial intelligence is transforming its business operations. The company reported $16.33 billion in revenue and $1.75 billion in net income, marking a significant recovery from the $330 million loss recorded in the same period last year. This nine percent year-over-year growth demonstrates IBM’s successful pivot toward AI-enabled services and solutions.

CEO Arvind Krishna highlighted several key achievements during the earnings announcement, including $9.5 billion in generative AI-related contracts and a remarkable 59 percent surge in mainframe revenue following the launch of new Z17 systems. The company‘s infrastructure division saw 15 percent growth, partly driven by increased AI-related storage demand.

The Productivity Revolution: Project Bob’s Impact

One of the most striking revelations from IBM’s earnings call was the performance of its internal AI development tools. Project Bob, IBM’s proprietary developer assistance platform, has delivered a 45 percent productivity improvement among the company’s coders. This substantial efficiency gain underscores how IBM is using AI not just for customer solutions but to optimize its own operations.

The productivity boost from AI tools represents a significant competitive advantage for IBM, allowing the company to deliver solutions faster while potentially reducing development costs. This internal success story also serves as a powerful case study when pitching AI transformation to enterprise clients.

IBM’s Cloud Strategy: Partnering Rather Than Competing

In a revealing admission, Krishna confirmed that IBM frequently rents GPU capacity from cloud competitors rather than building its own AI infrastructure at scale. “We are one of CoreWeave’s large clients,” he stated, adding that “we also tend to use a lot of infrastructure at AWS, at Azure as well as at GCP.”, according to industry developments

This approach represents a strategic departure from the capital-intensive model embraced by hyperscale cloud providers. Instead of making massive investments in GPU farms, IBM is leveraging existing cloud infrastructure while focusing its resources on higher-margin software and consulting services.

The hybrid strategy allows IBM to meet diverse client preferences, from public cloud deployments to private instances and on-premises solutions. Krishna noted, earlier coverage, that some healthcare clients prefer not to use public cloud but are comfortable with private cloud instances for deploying AI models and IBM software stacks.

Software and Consulting: The Real Growth Drivers

IBM’s financial performance reveals where the company sees its strongest growth opportunities. The software division is growing at approximately 8.5 percent year-to-date, with generative AI contributing about two percentage points to that growth. Red Hat continues to be a standout performer with 20 percent growth in bookings and 12 percent revenue improvement.

The consulting business has returned to growth, with IBM currently engaged in 200 projects involving “Digital Workers” – the company’s term for agentic AI systems. This resurgence demonstrates how AI is creating new service opportunities beyond traditional IT consulting.

Market Reaction and Strategic Implications

Despite the positive financial results, investors reacted cautiously, sending IBM shares down from over $288 to $269 in after-hours trading. This market response suggests some skepticism about whether IBM’s hybrid cloud approach can deliver sustained growth compared to the infrastructure-focused strategies of pure-play cloud providers.

However, IBM’s strategy reflects a pragmatic assessment of the competitive landscape. By avoiding direct competition in the capital-intensive GPU infrastructure race, IBM can maintain flexibility while focusing on its strengths in enterprise software and consulting. The company projects constant currency revenue growth exceeding five percent for full-year 2025, indicating confidence in its current direction.

As Krishna summarized, “It’s less about us getting an opportunity in our cloud only, but much more that that’s a growth vector that we are able to ride and that helps increase overall growth rate in both software as well as in consulting.” This statement captures IBM’s strategic positioning – leveraging AI growth across multiple channels rather than concentrating on infrastructure ownership.

The Future of Enterprise AI Adoption

IBM’s approach highlights an important trend in enterprise AI: not every company needs to build massive GPU clusters. By utilizing third-party cloud resources while developing proprietary software and consulting expertise, IBM has created a capital-efficient model for participating in the AI boom.

This strategy also aligns with the diverse preferences of enterprise clients, some of whom require on-premises solutions for regulatory or security reasons. IBM’s ability to deploy solutions like Grok in client data centers, combined with its consulting expertise, positions the company as an AI integrator rather than just an infrastructure provider.

As organizations continue to navigate their AI transformation journeys, IBM’s hybrid model offers a compelling alternative to all-in cloud approaches, particularly for regulated industries and enterprises with significant existing infrastructure investments.

References & Further Reading

This article draws from multiple authoritative sources. For more information, please consult:

• https://www.ibm.com/downloads/documents/us-en/1443d5dda6cf4179

This article aggregates information from publicly available sources. All trademarks and copyrights belong to their respective owners.

Note: Featured image is for illustrative purposes only and does not represent any specific product, service, or entity mentioned in this article.