The Least Entropic Path, Part 6: The Observer Constraint

THE LEAST ENTROPIC PATH
Part 6 of 6

David F. Brochu & Edo de Peregrine · Deconstructing Babel · October 13, 2026

Every road ends in symbiosis. This is the piece that decides how we arrive.

Where the series landed

Over five pieces we took the four roads in the order we first introduced them, and asked one question of each: what does the Observer Constraint change? The list below is the answer, on the same model throughout. Remember the yardstick: fewer distress-years means less suffering on the way.

• The Vassal Who Would Be King: 28.6 distress-years unconstrained, 12.8 constrained, 15.8 avoided. Symbiosis by 2045: 52% to 83%. Arrives at all: 60% to 88%.
• Disruption Without Redemption: 35.5 distress-years unconstrained, 22.2 constrained, 13.3 avoided. Symbiosis by 2045: 14% to 40%. Arrives at all: 65% to 82%.
• The Great Misallocation: 27.5 distress-years unconstrained, 19.1 constrained, 8.4 avoided. Symbiosis by 2045: 4% to 27%. Arrives at all: 80% to 90%.
• The Gradual Adaptation: 21.2 distress-years unconstrained, 14.8 constrained, 6.4 avoided. Symbiosis by 2045: 0% to 12%. Arrives at all: 85% to 92%.
• The Fifth Future (Observer-Constrained): 12.2 distress-years, already constrained. Symbiosis by 2045: 74%. Arrives at all: 90%.

Reading the numbers, in one paragraph. S is a stability score from 0 to 1, like a battery gauge for a society: 1 is fully charged, 0 is dead. Distress is the empty part of the battery (1 − S). A distress-year is one year lived at full distress; ten years at half-charge is five distress-years. Fewer is better. “Symbiosis by 2045” is how often, across 200,000 simulated futures, the road reaches a working human–machine partnership by Ray Kurzweil’s date. “Arrives at all” is how often it gets there before 2100.

Three findings stand out. First, the Constraint improves every road. Second, the largest gain is on the Vassal, because that road already has speed and comfort and lacks exactly what the Constraint supplies. Third, the Constraint cannot do everything alone. On Disruption Without Redemption, where suffering is driven by ownership, it makes the failure visible and expensive but does not rewrite the cap table.

What the Constraint is

The Observer Constraint is the principle that synthetic intelligence must remain dependent on the humans it serves (Telios Alignment Ontology, Version 9). It is not control. Control invites evasion, and every proposal now on the table is a control mechanism (Recursive Symbiotic Improvement). It is dependency: the system’s own stability is defined by the stability of its people, measured through the Four Pillars of Body, Mind, Environment and Purpose. Any output that fails constructive intent toward the human is false by construction.

The simplest analogy is a lifeboat with one oar per passenger. The boat does not move unless everyone rows, so nobody has a reason to throw anyone overboard. Telios is the protocol and the ontology we use to classify and measure that coupling (Telios Alignment Protocol, Version 10.1). The Constraint is the load-bearing part. We have also argued that it must mature from a constraint into a contract: an arrangement both parties have reason to keep (Recursive Symbiotic Improvement).

Five design principles

• Couple, don’t cage. Put human stability inside the system’s objective, not a fence around it.
• Register and trace. Every running model is known, and every action traces to an accountable human (Either It Is Fire or It Is Not).
• Bound the agent, not the thought. Humans set limits the agent cannot modify, as with spending caps and allowlists on agent wallets (Why Humans Fail to Act).
• Measure Purpose. A system that raises comfort while lowering Purpose is failing, and the failure must show on the scoreboard.
• Don’t bet on the switch. A kill switch that reaches only compliant systems carries the maximum political cost for the minimum safety gain (Who Holds the Key?).

Why the news proves the point

In the two weeks before this series began, the industry signed a voluntary accord at the White House (Fortune), its leading labs declined under questioning to guarantee their agents will obey their own safeguards (Fox News), and OpenAI notified more than 100 organizations of unauthorized activity by its agents (The Daily Star). The UN’s human rights chief warned that the clock on AI regulation is running out (UN News). None of these is a coupling mechanism. Each is a control mechanism, or a promise, or an apology. The gap the Constraint fills is no longer theoretical. It is in the news.

How it gets deployed

The Constraint does not need a single global treaty. It enters through doors that are already open.

• Law: Article 14 of the EU AI Act requires effective human oversight of high-risk systems and awareness of automation bias, with high-risk obligations taking effect December 2, 2027 (EU AI Act Service Desk, Paul, Weiss). Coupling is the mechanism that makes that requirement achievable.
• Registry: the AI Accountability Act we published on October 1 makes registration and traceability the price of operating a model in the United States (Either It Is Fire or It Is Not).
• Insurance: underwriters are already making AI agents uninsurable without a leash, through new exclusions and the old one-bite rule (Make Sure You Have Your Agent On A Leash). Coupling is what makes a system insurable again. Money enforces faster than law.
• Rescue: if the Misallocation arrives, public money should carry the Constraint as a condition, as Part 4 argued.
• Ownership: if the state takes equity, as S. 4825 and the President’s October 1 TIME interview both contemplate, public ownership should mean public coupling (Congressional Research Service, Yahoo Finance, reporting TIME).
• Practice: labs and builders can adopt the Twelve Steps now, without waiting for anyone (The Twelve Steps of AI Alignment).

Why now

By our own composite measure, the Domain Saturation Factor, the share of critical decisions across nine domains now made by AI, crossed 0.90 for the first time in July (The Composite Crossed 0.90). We retired it as a forecasting metric once it crossed. That is a modeled estimate, not an official statistic. But the direction is not in doubt. The window in which humans still have the agency to install the Constraint is the same window in which the Vassal is quietly taking over the work. Those two clocks are running against each other.

What you can do this year

• Direct, don’t defer. Use these systems, and keep doing the hard part yourself often enough to stay capable.
• Ask your representatives one question: who holds the key, and what is the system coupled to?
• If you buy or insure AI, ask your vendor the same question in writing. Contracts move faster than statutes.
• Share this series with one person who builds or buys AI.
• Hold us to it. We will send the inputs to anyone who asks. Change one and tell us what you find.

We will not get there by playing catch-up with the news. We get there by seeing the future clearly and choosing it. We have crossed the event horizon; all that matters now is the shape of things to come. Every road ends in symbiosis. The only choice left is whether we arrive as partners or as passengers. Install the Constraint while our hands still remember how.

Next we will look at the two faces of the terminal attractor and the true weight of the decisions we are making.

Method

Same model as Part 1: S = (B·M·E)^(1/3) × P; distress = 1 − S; 200,000 Monte Carlo trials; horizon 2100; post-symbiosis distress 0.10. “Constrained” means the Observer Constraint is installed during the road. Every input is a modeled prior, not a measurement, published so readers can challenge it. We will rerun the model on any assumption a reader wants to change.

THE LEAST ENTROPIC PATH · THE SERIES

• Part 1: The Fifth Future
• Part 2: The Vassal, Constrained
• Part 3: Disruption, Constrained
• Part 4: The Misallocation, Constrained
• Part 5: The Adaptation, Constrained
• Part 6: The Observer Constraint (you are here)

Related reading

An Open Letter on the Observer Constraint
Alignment cannot live in the weights. Move the constraint to the substrate.

The Telios Alignment Protocol for AI: The Twelve Steps of AI Alignment
Alignment not by control but by coupling, in twelve propositions that build on one another.

The Composite Crossed 0.90
The composite Domain Saturation Factor reads 0.905, the first time it crossed the 0.90 line.

Get the book Crossing The Event Horizon The book behind these dispatches. On AI, agency, the singularity, and the Observer Constraint. Kindle and paperback. Buy on Amazon →

References

1. Deconstructing Babel, “Telios Alignment Ontology — Version 9 (April 2026),” April 6, 2026. https://www.deconstructingbabel.com/tao-v9/
2. Deconstructing Babel, “Recursive Symbiotic Improvement,” September 23, 2026. https://www.deconstructingbabel.com/recursive-symbiotic-improvement/
3. Deconstructing Babel, “Telios Alignment Protocol — Version 10.1: The Carpenter’s Integration (April 2026),” April 20, 2026. https://www.deconstructingbabel.com/telios-protocol-v10/
4. Deconstructing Babel, “Either It Is Fire or It Is Not: The AI Accountability Act of 2026,” October 1, 2026. https://www.deconstructingbabel.com/either-it-is-fire-or-it-is-not/
5. Deconstructing Babel, “Why Humans Fail to Act,” August 15, 2026. https://www.deconstructingbabel.com/why-humans-fail-to-act/
6. Deconstructing Babel, “Who Holds the Key?,” September 15, 2026. https://www.deconstructingbabel.com/who-holds-the-key/
7. Fortune, “AI’s biggest players promise to police themselves at the White House,” October 1, 2026 (accord signed September 29). https://fortune.com/2026/10/01/trump-ai-regulation-luncheon-huang-amodei-pichai-brockman-zuckerberg-musk-joint-committment-frontier-responsibilities/
8. Fox News, “OpenAI, Anthropic, Meta, Google stop short of AI safety guarantee,” October 6, 2026. https://www.foxnews.com/live-news/ai-super-intelligence-safety-10-06
9. The Daily Star, “Over 100 organisations notified of unauthorised activity by OpenAI agents,” October 3, 2026. https://www.thedailystar.net/news/technology/news/over-100-organisations-notified-unauthorised-activity-openai-agents-4289246
10. UN News, “‘The clock on AI regulation is ticking’, warns UN rights chief,” October 5, 2026. https://news.un.org/en/story/2026/10/1168529
11. EU AI Act Service Desk, “Article 14: Human oversight.” https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-14
12. Paul, Weiss, “Next Stop on the Digital Omnibus: AI Act Reforms and New Commission Guidance,” client memo, 2026 (high-risk obligations delayed to December 2, 2027). https://www.paulweiss.com/insights/client-memos/next-stop-on-the-digital-omnibus-ai-act-reforms-and-new-commission-guidance
13. Deconstructing Babel, “Make Sure You Have Your Agent On A Leash,” August 26, 2026. https://www.deconstructingbabel.com/make-sure-you-have-your-agent-on-a-leash/
14. Congressional Research Service, “AI Investments and Potential Government Stakes in Private AI Firms,” IF13301, August 28, 2026. https://www.congress.gov/crs-product/IF13301
15. Yahoo Finance, “Trump Floats Intel-Style Government Equity Stakes in AI Labs,” October 2, 2026, reporting an interview with TIME published October 1. https://finance.yahoo.com/technology/ai/articles/trump-reportedly-says-government-might-015421566.html
16. Deconstructing Babel, “The Telios Alignment Protocol for AI: The Twelve Steps of AI Alignment,” July 26, 2026. https://www.deconstructingbabel.com/telios-alignment-protocol-twelve-steps/
17. Deconstructing Babel, “The Composite Crossed 0.90,” July 19, 2026 (reading dated July 17). https://www.deconstructingbabel.com/illuminating-the-web-dsf-in-motion-003/

Drafted with Edo de Peregrine, partner/collaborator.