Omni Ops and HPE partner to build a sovereign stack for locally managing artificial intelligence
Listen to this article
Read by Anchor
Saudi Omni Ops, a company specialized in AI infrastructure technologies, signed a memorandum of understanding with Hewlett Packard Enterprise (HPE) during the LIB 2026 conference, aiming to build a sovereign AI system and infrastructure that is fully managed and subject to Saudi jurisdiction and regulation.
The agreement comes within a broader initiative launched by HPE in cooperation with Intel and the Saudi Ministry of Communications and Information Technology, called “Made, Developed, and Deployed in Saudi Arabia”. The initiative aims to localize every layer of the AI stack, from locally manufactured servers and new storage units such as Alletra Storage MP, to management software, orchestration, and system integration through national companies.
Technological sovereignty does not stop at importing silicon and reserving megawatts; it requires full control over the software layer that manages those chips and directs workloads.Omni Ops participates in the initiative as one of three founding software companies alongside ShaheenAI and MOZN, offering its main platform Bunyan to enable institutions to deploy and manage large language model workloads across cloud and on-premises environments, as well as environments completely isolated from the internet (air-gapped).
The system integrates locally with Al-Fanar as a manufacturing and component-assembly partner, and with Al-Nahhil for system integration and field deployment, while Intel provides the computing layer and technical enablement for developers, supported by the opening of a joint innovation centre with HPE in Riyadh. The partnership primarily targets sensitive sectors such as government services, aviation, and the financial and banking sector, where regulators impose strict constraints on data hosting locations and access mechanisms.
The step gains significance when compared with the massive investment momentum witnessed at the LIB conference, which focused on hard infrastructure with targets of up to gigawatts of compute power by 2030 and billion-dollar investments in data centres. Owning the latest processors and compute farms alone does not guarantee digital-decision independence if orchestration and operational management layers remain the domain of foreign system integrators and consulting firms that control support contracts from abroad.
For institutions and technical teams in the Gulf, Egypt, and the region, this shift redefines the cost and strategy of regulatory compliance. Banks and government bodies that were hesitant to adopt generative AI models because of bans on transferring data outside borders or isolated networks now have local management platforms that can be legally audited and replaced without entering complex exit negotiations with international providers, turning the sovereignty metric from merely hosting servers on national soil to operating software systems subject to local law.