AIWiki
Malaysia
Back to all articles
Tools & Platformsoracleocicloud

Oracle AI

4 min readUpdated October 2026
Oracle AI
Type
Enterprise AI and cloud-infrastructure business
Parent
Oracle Corporation (founded 1977)
Core platform
Oracle Cloud Infrastructure (OCI)
Landmark deal
4.5 GW Stargate capacity agreement with OpenAI (2025)
Malaysia
US$6.5B announced investment; Malaysia West 2 (Kulai) region live Feb 2026
Related
OpenAI, Stargate Project, AWS, Microsoft AI
Oracle AI refers to the artificial-intelligence products and strategy of Oracle Corporation, the American enterprise-software and database company founded in 1977. Oracle's AI play rests on two pillars: Oracle Cloud Infrastructure (OCI), which supplies large-scale GPU capacity for training and serving AI models, and AI features woven into Oracle's database and enterprise applications. Its data-centre partnership with OpenAI has made Oracle one of the largest physical builders of AI infrastructure in the world.[1]

History and technology

Oracle entered the generative-AI era through infrastructure rather than consumer models. OCI offers managed generative-AI services running open models alongside GPU compute families designed for training and high-throughput inference, and the company has positioned its AI Infrastructure line — built on NVIDIA systems, with the first NVIDIA GB200 racks delivered to the Abilene, Texas Stargate campus in June 2025 — as the backbone for frontier-model training.[2]

The defining transaction came in July 2025, when OpenAI and Oracle announced an agreement to develop 4.5 gigawatts of additional Stargate data-centre capacity, reported as roughly US$300 billion of business over five years.[1] In September 2025, OpenAI, Oracle and SoftBank announced five new Stargate sites, pushing the programme toward its US$500 billion, 10-gigawatt commitment; the Abilene flagship runs on OCI.[2][3] Oracle's CEO at the time described the goal as delivering "the most performant and cost-effective AI training and inferencing" at unmatched pace.[3]

Oracle's second pillar is its installed base: OCI carries Oracle's own database workloads, and integrations such as Oracle Database deployments inside Microsoft Azure and Google Cloud pull enterprise data into AI-ready form without full migration. This makes Oracle a common boring middle in AI architectures — storage, retrieval and serving rather than the model itself.[4]

Why it matters for AI

Oracle demonstrates that in the current AI build-out, power and land matter as much as software: the company's jump from distant fourth-place cloud to hyper-scale AI builder came from signing capacity deals faster than larger rivals. It also concentrates risk — a large share of Oracle's growth now depends on a small number of enormous customers, OpenAI chief among them, making the Oracle-OpenAI relationship a systemic dependency in the AI infrastructure map.[1][4]

>See Also

🇲🇾Malaysian Context

🇲🇾 Relevance to Malaysia: Oracle is directly present in the country. In October 2024 it announced more than US$6.5 billion in AI and cloud investment in Malaysia, promising a local cloud region so customers could keep workloads and personal data in-country under the Personal Data Protection Act.[5] The Malaysia West 2 (Kulai) OCI region was declared available on 2 February 2026, sited in Johor alongside the state's wider data-centre corridor.[6]

For Malaysian enterprises — banks, insurers, palm-oil and manufacturing groups already running Oracle databases — the local region removes the cross-border latency and data-residency objections that previously kept those workloads in Singapore or the United States, and brings OCI's NVIDIA GPU capacity within national borders. It also intensifies competition in Malaysia's cloud market against AWS (ap-southeast-5), Azure, Google Cloud and local players such as YTL AI Cloud, giving National AI Office and MDEC-driven adoption programmes a domestic-option benchmark against which to negotiate. Malaysia's status as an OCI build site also feeds the broader question the sector is debating: whether data centres' power and water draw in Johor is justified by the investment and capability they anchor.

References

  1. ↑OpenAI. (2025). Stargate advances with 4.5 GW partnership with Oracle. https://openai.com/index/stargate-advances-with-partnership-with-oracle/
  2. ↑Data Center Frontier. (2025). OpenAI and Oracle's $300B Stargate Deal: Building AI's National-Scale Infrastructure. https://www.datacenterfrontier.com/machine-learning/article/55316610/openai-and-oracles-300b-stargate-deal-building-ais-national-scale-infrastructure
  3. ↑Reuters / SoftBank Group. (2025). OpenAI, Oracle, SoftBank plan five new AI data centers for $500 billion Stargate. https://group.softbank/en/news/press/20250924
  4. ↑Reuters. (2025). OpenAI, Oracle, SoftBank plan five new AI data centers, $500 billion Stargate. https://www.reuters.com/business/media-telecom/openai-oracle-softbank-plan-five-new-ai-data-centers-500-billion-stargate-2025-09-23/
  5. ↑Oracle. (2024). Oracle to invest more than US$6.5 billion in AI and cloud computing in Malaysia. https://www.oracle.com/asean/news/announcement/oracle-to-invest-in-ai-and-cloud-computing-in-malaysia-2024-10-02/
  6. ↑Oracle. (2026). New Region in Kulai, Malaysia — Malaysia West 2 now available, 2 February 2026. https://docs.oracle.com/iaas/releasenotes/oci/new-region-malaya-2.htm
Back to all articles