Could RoboCast Help Solve the AI Crisis? How?
Related Intelligence (RI) is non-neural generative intelligence. It builds each result by clear-box composition: parts drawn from addressed sources, arranged by a script, performed by role and approved at human gates. Every step is traceable, and every approved workflow is repeatable.
The Future, Present, and Past
of RoboCast
Three tenses. The explainer shows where Related Intelligence is headed. A recorded research session shows the work being done today, pushback included. And the patent’s own FIG. 7, animated, shows the RoboSurf screen designed in the mid-1990s, years before AI agents made news.
Related Intelligence, explained in about four minutes.
The oversight gap in autonomous AI, the patent’s own words, Human Moderated Recursive Improvement (HMRI), and three ways to take part. Read the claims it quotes →Can RI Dirigents (RoboCast Agents) Make High-Stakes AI Safer?
Damon C. Torres and Marcy J. Gordon give a search-based AI the issued RoboCast claims and push back when it drifts. This is where the research ended, read aloud. The AI’s answers are AI-generated; check them against the patents yourself. Watch the full research session →
{{f7cap}}
The RoboSurf screen, as the patent drew it
FIG. 7 of US 7,155,451 shows the RoboSurf screen: Search, Pick, Gather, Schedule and Play, with the schedule, controls and content window around them. Step through it item by item in the Patent Hub, in the patent’s own words. Watch FIG. 7 animated → See all the animated figures →A different kind of digital intelligence.
Clear-box composition
Neural AI generates by black-box approximation: no one can record what happens inside it. RI generates by clear-box composition. Each performer’s role, source and decision is addressed and logged, so every RI step can be traced and replayed.
Why clear-box composition matters →The public-domain advantage
The RoboCast architecture is in the public domain. Researchers, builders and policymakers can study it, test it and build on it, and the record of all four builds is open in the Archive.
Explore the Archive →Generative Related Intelligence
AI co-writes the scripts. RI composes from addressed parts and runs each approved workflow step for step. People approve at every human gate. Together: AI + RI + Human Workers = Safer SuperIntelligences.
How AI and RI work together →Privacy you can see
Much of the web records you to sell your attention. RoboCast keeps a record that works for you: AI writes the instructions without seeing your personal information, RI performers touch only the data each step names, and the record of what they did can be shown to you and erased when you ask.
How RoboCast protects privacy →Conservation, Control, and Collaboration: A Blueprint for Generative Related Intelligence
A blueprint for developers of all sizes building generative RI apps and Dirigents. It defines clear-box, compositional and combinatorial as design properties, shows how RoboCast realizes each across four builds, and maps the pattern onto research reports, code, images, audio, video and XR.
It closes with an invitation: a race to the top, with builders competing openly to do better on conservation, control and collaboration, and sandboxing at scale for the surprising behavior Dirigents can produce.
The paper goes to builders on request. Building something now? Write to alliances@robocast.com.
“RoboCast's RI architecture is built to address three of the biggest problems facing AI today: conservation, control, and collaboration. It uses less energy and compute, keeps humans firmly in charge, and lets intelligent systems work together rather than in silos. That combination is why I've stayed involved as an investor and board member.”
Every RoboCast runs the same five steps — and keeps the receipts.
RoboCastOS decides whether each task goes to AI, RI or both, like two hemispheres of one brain working hand in hand.
- S · 01SearchRun an online search and turn each result into an addressable node.
- G · 02GatherCollect each node’s resource: a web page, a file, an app or another tool’s output.
- P · 03PickKeep or skip each node; every choice is stored and shapes the next show.
- S · 04ScheduleArrange the picks into a show: in sequence, branching, or side by side.
- P · 05PlayRun the show with pause, rewind and fast-forward, and record every step.
Three ways in.
Read the claims. Watch the figures move.
US 7,155,451 and four continuations make up the five-patent RoboCast portfolio, invented by Damon C. Torres. The Hub reproduces the issued claims verbatim, alongside the patent’s own figures, animated step by step.
Running in production years before the rest of the field arrived.
“I grew up with Damon, who stood out even then. He began coding at nine, and at 19 we helped him start his first company, ExpoVideo ShowTapes. In 1996, after building and selling PC Expo along with hundreds of other trade events, I became Robocast's first angel investor.
Damon built a different kind of digital intelligence. It isn't neural-network AI. Every connection is defined, every action is traceable, and every rule can be read by the people who govern it, at a fraction of AI's cost. That lets Robocast's Related Intelligence (RI) handle the work that demands structure, accountability, and an audit trail, freeing today's AI for the improvisational work it does best, so companies profit sooner.”
Pre-register for the Training Cohort
Learn the architecture behind Related Intelligence and Dirigentic systems. Pre-registration is free and commits you to nothing. Testing Group sessions with the RoboCast team are by invitation.
Hear what an AI made of the claims
This 15-minute audio explainer was generated with Google NotebookLM in September 2025. It shows what one AI tool produced from the issued RoboCast claims, the right questions, enough context and one more ingredient: our list of common wrong conclusions about RoboCast, with the pushback to each. The pushback is where the most useful insights show up.
AI-generated audio. AI can make mistakes, and this is not a statement by Robocast Inc. or Damon C. Torres, or an interpretation of the claims. The issued claims are reproduced verbatim in the Patent Hub.