Screen to Software, Silicon and Robotics: Vatsal Soin Files 0 to 1 Invention Patents for Superintelligence-Scale AI Systems
AI agents now write software, run machines and depend on silicon in data centres that draw gigawatts of power. Inventor Vatsal Soin's 0→1 Doctrine is designed to check each AI action routed through it on one scale from 0 to 1 before it runs, in software or in silicon, while Band-Based Archive Triage (BBAT) releases only signals from stored data, through secure hardware. Each filed invention works alone or with the other.
|
KEY POINTS Two filed inventions address the moment AI leaves the screen. The 0→1 Doctrine places each limit, from a robot's load to a contract's cost, on a 0 to 1 scale against its own range and checks it before AI acts. BBAT sorts stored data by its labels and releases only derived ranges, through secure hardware. Both are designed to work with the systems organisations already run. |
AI HAS LEFT THE SCREEN
Artificial intelligence no longer lives only in chat windows. AI agents place orders, book services and write software. Robots move stock through warehouses, and the chips that run them fill data centres. Each action now touches two layers: the hardware that carries it out and the software rules that should govern it. The patents filed by inventor Vatsal Soin address both layers.
WHY THE CHECK MUST COME FIRST
Much AI oversight still works after the fact, through logs, audits and recalls. When an agent can commit money or move a machine in a fraction of a second, a fix afterwards can arrive too late. The 0→1 Doctrine is designed to place the check before the act, so an action proceeds only when what it needs fits what is allowed.
ONE SCALE FOR EVERY UNIT
The Doctrine places any measurable value, whatever its unit, on a scale from 0 to 1 against its own range. A machine's force limit, a contract's cost limit and a region's legal rule are each tested on their own terms; a yes-or-no rule must simply pass. If any fails, the action can be changed, rerouted, held or stopped before it runs.
What changes: limits in different units are checked together.
Figure 1. Two layers, one scale: each routed AI action is checked against its own limits before it runs.
IN SOFTWARE OR IN SILICON
The filing describes the Doctrine running as software, firmware, hardware or a mix, on ordinary computers, AI accelerators and platforms built for quantum-resilient security. A chipmaker, a cloud provider or a factory can choose where the check lives, and a check held in hardware is harder for software to bypass.
What changes: the same rule can sit inside an app, a controller or a chip.
NO RIP AND REPLACE
Safety plans often stall when they demand new systems. The Doctrine is designed to turn its results into signals that older business, factory and control systems can read, through adapter layers.
What changes: protection can arrive without a rebuild.
DATA STAYS WHERE IT LIVES
Where configured, the Doctrine computes on the device or site that holds the data and shares only a bounded signal, while the original record stays with its lawful holder.
AGENTS, SWARMS AND ROBOTS
The Doctrine describes coordination among human workers, software agents and robotic systems. Its Orchestration and Compliance Token (OCT) is designed to pause, reroute, reassign or escalate work while it runs. A governed action that runs ends in an Actuation Compliance Receipt (ACR), a sealed record of what was checked, which an auditor can verify later.
ONE ORDER, START TO FINISH
Illustrative example: an AI agent sends an urgent order to warehouse robots. Before any robot moves, the Doctrine checks how much each one can carry, how much battery it has left, the delivery deadline and the safety rules. One robot is almost at its limit, so the job goes to another robot that passes every check. The warehouse's existing stock software is updated through a connector, and a receipt records why.
Figure 2. One urgent order, from an AI agent to warehouse robots and the existing stock software.
WHEN THE NETWORK GOES DOWN
Factories, farms and clinics do not always have a connection. The Doctrine's filing describes offline tokens carried by QR code, text, voice or paper and reconciled when the connection returns.
What changes: checks can run locally where networks do not.
RECORDS BUILT FOR THE QUANTUM ERA
Future quantum computers could break much of today's encryption. The Doctrine's receipts are designed to use upgradable, quantum-resilient signing, so a record made today can still be checked for tampering years from now.
PEOPLE SET THE LIMITS
Humans set every range. In the Doctrine, people can approve, adjust or halt decisions.
HARDWARE THAT HOLDS THE DOOR
The 0→1 Doctrine checks what AI does. The second filed invention, Band-Based Archive Triage (BBAT), governs the stored data AI learns from. Through a metadata interface, records are sorted by their labels, such as age, category and legal hold, and marked keep, delete or promote, so an archive need not move. Promoted records are worked on only inside a Trusted Execution Environment (TEE), a protected area of a chip. The exit stays shut until a Hardware Attestation Certificate (HAC) attests that the working copy was deleted under Delete-Before-Share. Groups too small are refused and each range is tested against outside data before release. Held or doubtful records go through a Human Override Pathway (HOP) to an authorised reviewer, who decides from the labels without seeing the content.
Figure 3. BBAT's hardware door: sorted by labels, worked on inside the chip, released only after deletion is attested.
TOWARD SUPERINTELLIGENCE
Artificial general intelligence (AGI) and superintelligence do not exist today, and nothing here claims to solve them. Some researchers foresee a Singularity, when AI improves itself faster than people can follow. Even then, AI will likely run on chips and act through code. Both inventions are designed for that path: a check before the act, and data shared only as checked ranges.
WHAT ADOPTION TAKES
An organisation connects its systems, maps its limits and sets its ranges; results then depend on those ranges and on accurate data. Each invention has its own filing and works alone.
QUESTIONS READERS ASK
|
01 Does it need special chips? Not for the Doctrine. BBAT's secure release uses a protected area built into a chip. |
02 Do we have to change our AI? No. The inventions are designed to work with the AI already in use. |
|
03 What if a required input is missing? The Doctrine is designed to send the job back for redesign or another supplier. |
04 Who can check a decision later? An authorised auditor, using the sealed receipt. |
CLOSING NOTE
Code writes the instruction and silicon carries it out. Between the two, one scale from 0 to 1, with limits set by people, can decide whether an AI action should happen at all.
This can be tested, live, via API, governed against ungoverned, side by side.
THE INVENTOR
Systems theorist, inventor, and serial entrepreneur Vatsal Soin holds grants in the US, Japan, India and more; filings span six continents; latest grant, AI footwear, August 2026. Inventions: footwear, apparel fit, Sharable Size Cards, digital twins, robotics, global governed data, archive triage, quantum-resilient cryptography, AI governance. SIM–RMIT and NTU alumnus. Every prior invention was a step; 0→1 is the destination.
SELECTED REFERENCES
Granted: US Patent 12,446,652 B2 · Japan Patent 7560909 · India Patents 454081 and 599317. Filed: PCT/IN2025/051943 · US 19/489,595 · India 202511115781 · Australia AU2022450649 · India 202611113867 (23 September 2026) · PCT/IN2026/052042 (1 October 2026).
DISCLAIMER
Informational only. Not certified. No endorsement implied. Not investment advice. Examples are illustrative, not field results. Vatsal Soin · © 2026 All Rights Reserved.





