OrbitronAI has launched System 1 Models, an AI service hosted in Saudi Arabia and the UAE that keeps sensitive data within national borders.
The technology company said the service lets applications assess information, evaluate predefined outcomes and return structured decisions through an API. It runs on OrbitronAI’s regional infrastructure.
The name draws on psychologist Daniel Kahneman’s distinction between fast, intuitive System 1 thinking and slow, deliberate System 2 thinking. Chat-based assistants typically generate open-ended responses, while System 1 Models are designed to deliver fast, structured decisions.
Saulius Adomaitis, co-founder and chief executive of OrbitronAI, said: “AI adoption depends on how effectively its outputs drive decisions within real business workflows. With System 1 Models, we are extending our approach beyond conversational interactions, enabling applications to make structured decisions while addressing regional data sovereignty requirements. We see significant potential across enterprise applications where decisions must be fast, actionable and integrated directly into operational processes.”
Through the API, developers submit a description of a situation, such as a customer message, a transaction record or a maintenance log, along with the questions they need answered. Each question has a predefined set of outcomes the application is willing to accept. Rather than composing free text, the model weighs the input against those outcomes and assigns a probability to each.
The service offers three types of output: a binary question that returns the probability of a yes answer; a choice among predefined options, such as which team should handle a request; and a score on an ordered scale, such as a risk level from low to critical.
OrbitronAI said the models support classification, ranking, risk assessment and intent detection, and are aimed at time-sensitive applications that need an answer within the workflow.
The company said data localisation requirements leave organisations in regulated industries choosing between widely used AI models served from overseas data centres and keeping sensitive data within national borders while limiting automation. It said System 1 Models run on its own infrastructure inside Saudi Arabia and the UAE, allowing developers in banking, government, energy, aviation and other regulated sectors to build applications on data that never leaves the country.
Ashu Gupta, co-founder and chief product and AI officer of OrbitronAI, said: “Most enterprise AI decisions don’t need a model that writes an essay. They need one that answers a narrow question, fast, with a result software can act on. System 1 models do that at a fraction of the cost and latency of a general-purpose LLM, because they don’t spend tokens generating prose nobody reads. It’s the part of the stack our customers’ compliance and operations teams have been asking us for. The real test is whether a regulated team can trust each decision enough to let it run without a human in every loop, while meeting data localization requirements.”
Organisations interested in early API access can join a waitlist, stating their preferred hosting region, Saudi Arabia or the UAE, and their expected monthly API call volume.
System 1 Models build on NovaOS, OrbitronAI’s agent orchestration and governance platform, introduced in October 2025. NovaOS is designed to let organisations in aerospace, energy, government and industrial sectors deploy and manage AI agents within existing systems, in cloud, on-premise or hybrid environments.
OrbitronAI was founded in 2025 by Saulius Adomaitis, Ashu Gupta, Luvy Singh, and Poonam Gupta. It operates in four regions: the USA, MENA, Europe, and Asia.
Read next: Costain joins alliance to hire care leavers
Are you a building professional? Sign up for a FREE MEMBERSHIP to upload news stories, post job vacancies, and connect with colleagues on our secure social feed.


