What we do
Oasis-X is an engineering and research company. We build the control, safety, and measurement layers for systems that act without a person in the loop: software agents, industrial plants, and energy and climate infrastructure. Our customers are the operators and communities who pay when those systems fail.
Symbiotic intelligence is our standard for that work. The machine and the person each do what they do well. The person keeps the authority to decide.
Who benefits
-
People subject to automated decisions
A person keeps the authority to review, approve, and reverse any action an AI system takes. We build the approval gate and the record that make this possible.
-
Small operators, growers, and independent engineering teams
We publish our control, coordination, and security software as open source. A small operator runs and inspects the same tools a large firm uses, at no license cost.
-
Communities exposed to industrial and climate harm
We produce independent measurement and evidence. A property owner or a community uses that evidence in an insurance claim or a legal claim.
-
Industrial operators and their neighbors
We redesign processes to use less energy, less water, and less material. This lowers cost for the operator. It lowers emissions and waste for the community around it.
Oasis-X is the trade name of Open Agriculture Systems & Integrated Supply Inc., a Delaware public benefit corporation. The four groups above are the beneficiaries we are accountable to.
What we build
The three products below are our AI work. We release all three as open source. Our practice covers three further areas: industrial process design, climate and energy modeling, and product development. Products & Services lists all four.
01
Agent Governance
Teams that run AI agents lose the record of who decided what. We keep every claim, handoff, and decision in version control. A person reads the record and reverses a bad decision.
02
Agent Security & Privacy
Agentic AI usually ships without a safety layer. We supply one. Credentials stay outside the model. A person approves each sensitive action. Every action leaves a record. An attacker cannot use an agent to reach a customer's data or money.
03
Alignment & Evaluation
Human values do not reduce to one score. We measure the trade-offs between helpfulness, safety, honesty, privacy, and agency. The people an AI system affects can then see which value the system gave up.
Principles
Global AI Challenges
AI systems have to do more than perform tasks. They have to protect privacy, respect agency, and stay governable across their lifecycle. Researchers at the University of Central Florida named six such challenges. Our answer to each one follows.
| Challenge | Oasis-X response |
|---|---|
| Human Well-Being | Workflows that preserve context across sessions and surface conflicts in human preferences instead of averaging them away. |
| Responsible AI | Multi-objective value modeling, geometric safety barriers, and supply-chain audit on every installed skill. |
| Privacy | Plaintext credentials never enter the LLM context. Local-first architecture. Signed logs for prompt-injection incidents. |
| Human-Centered Design | A public risk ladder. Approval gates in the gateway. Explicit lifecycle controls: an agent can sleep, compact, and hand off. |
| Governance & Oversight | Git-native traces, signed logs, and lifecycle artifacts. We run our own development on the protocol we ship. |
| Human-AI Interaction | Inspectable, interruptible, and accountable. Operator approval runs through an explicit gate, and transcripts are append-only. |
AI Risk Ladder
- 01 Assistive AI drafts, summarizes, suggests, or explains. Human decides.
- 02 Delegated AI performs bounded tasks with reviewable outputs and logs.
- 03 Operational AI can affect systems, customers, money, or data. Requires approval gates, monitoring, and rollback.
- 04 High-Impact AI affects health, legal, finance, employment, or essential services. Requires formal governance.
- 05 Prohibited Deceptive, coercive, privacy-invasive, or designed to remove meaningful human control.
Work with us
Tell us the system you have to keep under control: an AI deployment, a plant, a site, or a claim. We will tell you what we can measure and what we can build.