newsfilter
← Latest News
Earnings Call Transcripts·

Oracle & IBM Power Present: Modernize for AI resilience and security while optimizing costs and existing infrastructure - ILOG SA

  • Financial Performance:

    • No specific financial metrics (revenue, earnings, margins, cash flow, or debt) were disclosed as the call focused on a strategic technical partnership and product roadmap rather than quarterly financial reporting.
    • Mention was made of significant joint investment, including "thousands of man-hours" and "millions of dollars in hardware and farm equipment" dedicated to the Oracle 26AI on IBM Power development.
  • Guidance and Future Outlook:

    • Oracle Database 26AI is scheduled for release on IBM Power and AIX platforms in the first half of 2027.
    • The extended timeline to 2027 is attributed to a deliberate focus on remediating thousands of vulnerabilities detected via foundation models prior to release, ensuring a secure and stable platform rather than a lack of engineering effort.
    • Long-term collaboration between IBM and Oracle is confirmed to extend well beyond the 2026/2027 release window, with both companies expressing strong commitment to the partnership.
    • IBM and Oracle plan to make current OpenSSL encryption improvements available for Oracle Database 19c as well as the upcoming 26AI release.
  • Business Segments and Product Lines:

    • IBM Power 11: Highlighted as the optimized platform for Oracle 26AI, featuring native autonomous operations, 6 nines (99.9999%) availability, and zero planned downtime for maintenance via live partition mobility.
    • Oracle Database 26AI: A major release introducing 300+ new capabilities, including native AI agents, vector search, JSON duality, and property graph support.
    • Key Product Features:
      • Autonomous Operations: Native AI agents for infrastructure management, automated patching, and 15x faster capacity resolution.
      • Security: "IBM Power Cyber Vault" for ransomware recovery in under a minute; full stack Post-Quantum Cryptography (PQC); immutable backups via ZDLRA.
      • Data Integration: Support for Apache Iceberg, federated queries across heterogeneous systems (SQL Server, Teradata), and a Mongo API for document data.
      • AI Agents: Capabilities for "Private Agent Factory" (on-prem) and "Select AI" (cloud), with agents running natively within the database for secure, constrained SQL generation.
  • Market and Competitive Landscape:

    • Differentiation: IBM positions Power 11 as a cohesive, vertically integrated platform (silicon to firmware) versus competitors' "cobbled-together" multi-vendor stacks, citing superior performance and reduced complexity.
    • Converged Platform Strategy: Oracle promotes 26AI as a single platform capable of handling relational, JSON, graph, and vector workloads, reducing data sprawl and the need for separate specialized databases.
    • Market Trends: Emphasis on "Agentic AI," the need to mitigate AI-driven security risks (vulnerabilities, guardrail bypasses), and the looming threat of quantum computing ("harvest now, decrypt later").
    • Quantum Readiness: Both companies highlight PQC adoption to address regulatory requirements (DORA, government procurement) and long-term data sovereignty.
  • Risks and Challenges:

    • Security Vulnerabilities: The discovery of widespread vulnerabilities via AI models has necessitated a delayed release to ensure all issues are patched before shipping.
    • Quantum Threats: Concerns regarding asymmetric cryptography (RSA, ECC) being vulnerable to future quantum decryption capabilities.
    • Ransomware: Addressed through automated detection (less than one minute) and "Cyber Vault" safeguards.
    • Operational Complexity: The shift to autonomous operations and AI agents requires new guardrails to prevent arbitrary SQL generation or unauthorized data access.
  • Management Commentary and Tone:

    • Tone: Highly collaborative, confident, and technical; characterized as a deep, 40-year partnership with a personal commitment from IBM CTO Satya Sharma ("26AI is my personal project").
    • Confidence: Strong assurance that 26AI features on IBM Power will be feature-parity with other Oracle platforms (Exadata), barring Exadata-specific storage indexing.
    • Strategic Focus: Emphasized "speed" in AI operations paired with "risk mitigation" as the core value proposition of the joint solution.
  • Analyst Questions and Answers:

    • Release Date: Confirmed for H1 2027; delay explained as necessary for comprehensive vulnerability remediation using AI testing.
    • Platform Support: Confirmed support for both Power 10 and Power 11, and AIX 7.2/7.3.
    • Feature Parity: All 26AI features are available on IBM Power, with the exception of storage-specific features native only to Exadata.
    • Hardware Requirements: IBM Spire cards are not required to utilize 26AI on Power; the standard MMA accelerators and SIMD capabilities on Power 11 will be utilized.
    • Cloud vs. On-Prem: Clarified that AI capabilities (like agents and vector search) are available on-premises via the DBMS Cloud package and Private Agent Factory, not limited to cloud-only services.
    • Future Roadmap: Reaffirmed long-term collaboration beyond 2026/2031.
  • Other Key Points:

    • Demo Focus: The session included plans to demonstrate "Enterprise Data to AI Power Decisions" and "Securing in the Quantum Era."
    • Open Source Contribution: IBM is contributing to open-source projects like OpenSSL and ONNX runtime with Power-optimized encryption and AI runtimes.
    • Actionable Steps: Customers are encouraged to follow a "checklist" to prepare: upgrade to Oracle 19C, update to AIX 7.2/7.3, apply security patches, and upgrade to Power 11.
    • Event Promotion: Both companies highlighted their presence at the upcoming Oracle AI World conference in Las Vegas.
Company
ILOG SA
Keywords
ILOG SA