AnswersPublished15 sources
360 Says Its Sovereign AI System Is Already in Use at Some Organizations
On October 23, 2026, 360 founder Zhou Hongyi said the company’s sovereign AI operating system was already in use at some government and enterprise organizations. Zhou connected DeepSeek’s open source DeepInfer framework with a wider case for deploying AI on an organization’s own servers, rather than relying entirely...
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer
Key takeaways
- On October 23, 2026, 360 founder Zhou Hongyi said the company’s sovereign AI operating system was already in use at some government and enterprise organizations.
- Zhou connected DeepSeek’s open source DeepInfer framework with a wider case for deploying AI on an organization’s own servers, rather than relying entirely on outside platforms.
- 360 says the system includes model and compute options, enterprise knowledge tools and agent security controls; available reporting describes the architecture but does not independently verify its performance or safeg...
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: What did 360 Security Technology founder Zhou Hongyi announce on October 23, 2026, about its sovereign AI operating system and its use by go. Article summary: On October 23, 2026, Zhou Hongyi said 360 had built a “sovereign AI operating system” and that it was already in use at some government and enterprise organizations. He presented it as a way for customers to retain contr. Topic tags: general, general web. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clic
On October 23, 2026, 360 founder Zhou Hongyi said the company had built a “sovereign AI operating system” that was already in use at some government and enterprise organizations. He framed the announcement around a central question for the agent era: can an organization control the AI infrastructure, data and accumulated business knowledge its agents depend on? The reports do not identify the customers or state how many deployments there are.1
2
Why Zhou connected DeepInfer to AI sovereignty
DeepSeek’s internal inference framework, DeepInfer, had recently been opened to the public. Reports describe it as optimizing computing schedules so CPUs and graphics processors can work concurrently, reducing waiting and repeated computation and helping the same hardware handle more tasks.1
2
Zhou characterized the open-source release as a sign of “technical confidence” and an inclusive approach to sharing open-source benefits. He also argued that more mature open models and inference software can make it possible for organizations to run AI on their own servers and build agents tailored to their business, without training a large model from scratch.1
2
In that sense, DeepInfer was part of a broader argument about who gets to control AI—not just a story about inference efficiency. Zhou’s definition of “sovereign AI” centers on the user having the leading role. He described five levels: national, institutional, enterprise, individual and family, and device. The available reports list those levels but do not give a separate technical definition for each.1
4
Why enterprise control matters in the agent era
Zhou warned that when companies rely on external AI services, providers may control the models, computing resources and platforms. In his account, this can expose businesses to price changes or service adjustments, while leaving them dependent on systems they do not control.1
2
He also raised a question of ownership: as agents handle business data and accumulate knowledge and experience, who controls those assets? For Zhou, enterprise AI sovereignty is therefore about more than choosing a model. It also concerns control of compute and data, and ownership of the knowledge built up through AI use.1
19
What 360 says its system includes
The described system brings together compute resources and scheduling, sovereign large models, enterprise AI knowledge bases, an agent factory, Agent Teams for human–agent collaboration, and security governance.2
19
According to reporting on the company’s description, organizations can choose models, schedule compute and use their own data for post-training. One account also says that data remains within the organization’s network and that accumulated knowledge, skills, experience and task records belong to the organization.2
3
19 These are reported product claims; the available material does not provide detailed technical or contractual terms for each form of ownership.
Agent permissions and human oversight
360’s system is described as controlling and auditing agent permissions, tool calls and execution, with people overseeing key operations.19 Those mechanisms are intended to keep agents within authorized limits and leave final control with the organization. The reporting describes the proposed governance approach, but does not independently demonstrate how effective the controls are in practice.
What the announcement does not establish
The reports say the system is in use at “some” government and enterprise organizations, but do not name those customers or provide deployment counts.1
2 They outline the system’s components and intended controls, but do not supply detailed implementation specifications or independent validation of its performance and security.
2
19
The announcement therefore signals 360’s approach to enterprise AI control, while leaving key questions about deployment scale and real-world effectiveness unanswered.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "360 Says Its Sovereign AI System Is Already in Use at Some Organizations"?
On October 23, 2026, 360 founder Zhou Hongyi said the company’s sovereign AI operating system was already in use at some government and enterprise organizations.
What are the key points to validate first?
On October 23, 2026, 360 founder Zhou Hongyi said the company’s sovereign AI operating system was already in use at some government and enterprise organizations. Zhou connected DeepSeek’s open source DeepInfer framework with a wider case for deploying AI on an organization’s own servers, rather than relying entirely on outside platforms.
What should I do next in practice?
360 says the system includes model and compute options, enterprise knowledge tools and agent security controls; available reporting describes the architecture but does not independently verify its performance or safeg...