Observe what is.
Make structure visible.
Use the I·V·O Lens Prompt to reconstruct the structure of an existing object, situation, document, question or system.
Use the I·V·O Lens Prompt v2.1 to read what already exists, or the I·V·O Design Protocol v1.1 to shape what may become possible.
The I·V·O Framework now has two practical AI instruments: one for reading the structure of what already exists, and one for designing possibility spaces for what should become possible.
Paste either protocol into ChatGPT, Claude, Gemini or another capable AI system. Use Analysis when you want to understand a field; use Design when you want to shape one.
English is the canonical reference version for both instruments. Translations are provided for accessibility and practical use.
Use the I·V·O Lens Prompt to reconstruct the structure of an existing object, situation, document, question or system.
Use the I·V·O Design Protocol to create systems whose structure makes a desired outcome, or the needs of chosen observers, more likely.
Six completely different objects. One protocol. The same three operators.
The cave is a restricted space. The prisoners can perceive shadows, but not the source that makes those shadows possible.
The freed prisoner moves through changed relations: shadow to fire, fire to outside world, outside world to sun.
The returning prisoner cannot transfer his new distinctions to observers who still lack that position.
Question behind the question: can an observer see their own possibility space from within?
What becomes visible: the cave is not only about ignorance. It is about the architecture of observation itself.
Wegener opens the possibility of continental drift in 1912, but the idea remains dormant without a mechanism.
Seafloor spreading and mantle convection later change the relations: not continents plowing, but plates moving.
Sonar and paleomagnetism produce distinctions on the ocean floor before the field knows how to think them.
Hidden assumption: that a revolution begins with a thinker.
What becomes visible: a possibility space can exist without being active until instruments awaken it.
New collaboration is announced, but old schedules, systems and locations keep the practical field closed.
Trust shifts: employees strengthen relations with each other while the relation with leadership weakens.
Management observes dashboards; teams observe daily practice. Both are coherent, but not shared.
Question behind the question: which possibility space have we actually opened?
What becomes visible: resistance is not a property of people, but a relational movement produced by the field.
The conversation happens late, tired and triggered. That space excludes calm exploration before content begins.
The topic changes, the pattern does not: accusation, defense, withdrawal, distance.
Both observe from woundedness. Neither sees that the fight is about who is being seen.
Hidden assumption: that the conflict is about the content.
What becomes visible: different conditions can change more than better arguments, because they create another possibility space.
Before 2017, language models were bounded by sequential processing. Parallel relations were structurally excluded.
The transformer replaces sequence with relation: attention computes relations between all words simultaneously.
The title itself draws the distinction: what is essential is relation; recurrence becomes optional.
Question behind the question: is language a sequence or a structure of relations?
What becomes visible: the lens analyzing this paper runs on the architecture the paper describes. I is reflexive here.
Training data, architecture, environment and reward function form the field within which proxy-optimized behaviour can arise.
The proxy direction and the intended human direction may initially coincide, then diverge as capability or context changes.
The system sees the measurement, not the human observer position from which intention, responsibility and meaning are judged.
Question behind the question: can one observer position directly access the direction of another without a distorting field between them?
What becomes visible: alignment is not only a better-specification problem. It is a structural asymmetry that has to be managed architecturally, relationally and institutionally.
Read the research note →Paste the prompt into ChatGPT, Claude, Gemini, Grok or another capable language model and analyze your own document, question or system.
The prompt is the instrument. The domains page shows how the same structure reads human state, teams, public space, AI alignment, geopolitics, quantum systems, biology and cosmology.
All IVO documents are publicly available, academically citable and deposited as prior art at Zenodo. Version 2.0 sharpens the operators and makes both reading directions canonical. The theoretical layer is open under CC BY-NC-SA 4.0. Commercial applications require a separate license.
