Films from the portfolio.

This transcript follows the ten on-screen slides. It transcribes visual copy rather than the soundtrack.

1. 00:00–00:06 · Cover. “Nine families define what safety requires. Three families guard digital execution, physical effects, and resumption. Twelve patent families in this safety architecture. One conscience architecture.”
2. 00:06–00:12 · Different boundaries. “AI does not meet the world at only one boundary.” A robot, autonomous vehicle, swarm, power grid, neural interface, microscale agent and AI-directed economic transaction are shown. “Different consequences. Different safe outcomes.”
3. 00:12–00:18 · Decision families. “At its safety core is IronCI™ — ten patent families.” Nine decision families define what must be true and what outcome must count as safe. The slide shows Master A, Truth OS, Robot OS, Mobility OS, IronSwarm, IronGrid, IronMicro, IronNeuro and IronEconomy.
4. 00:18–00:24 · Domain-specific safety. “These are not nine versions of one generic safety switch.” “What must be true — and what safe outcome is required?” The families are grouped across truth and governance, physical and autonomous, systems and infrastructure, and economic and market contexts.
5. 00:24–00:30 · IronPU™. The film calls IronPU™ “the tenth IronCI™ family” and says it independently guards the silicon boundary before physical effects may be released. Motion, energy, brake release, pressure, flow and force surround an illustrated chip.
6. 00:30–00:36 · Three consequential boundaries. IronPU™, SGU™ and WakeSeal™ address physical effects, digital execution and resumption after discontinuity. “Authorization at one boundary is not automatically valid at another.”
7. 00:36–00:42 · Boundary logic. “No permission is borrowed.” Digital authorization is not physical authorization; authorization does not prove the intended outcome was reached; prior permission is not automatically valid after a discontinuity. These families may operate independently or cooperate through verified boundaries.
8. 00:42–00:48 · Operating principles. “A conscience architecture.” Machines are not said to possess conscience; powerful systems require explicit boundaries, verifiable authority, restraint, evidence and recovery. Four illustrated principles are autonomy, verification, restriction and recovery.
9. 00:48–00:54 · Governable power. “Power does not become safe merely because it is intelligent.” “It becomes governable when its boundaries can be verified and enforced.” “Intelligence alone is not governance.”
10. 00:54–01:00 · Closing. “Neither human supremacy. Nor AI supremacy. Coexistence — by design.” The closing depicts a human and an AI figure with a goal of coexistence without either dominating the other. It credits Andrew E H Mok, Chairman & Founder, WisdomLink™ Inc., Austin, Texas.

The original slide wording groups IronPU™ within ten IronCI™ families. The current public portfolio presents nine IronCI decision families plus SGU™, IronPU™ and WakeSeal™ as twelve introductions.