Why institutions are shifting authority from human agents to executable grammars The Disappearance of the Agent For centuries, institutions were built on a simple premise: authority must be exercised by identifiable subjects. A law was signed by a monarch, a decree was announced by a minister, and an audit was validated by an official. Agency anchored responsibility, making it possible to trace every decision back to someone who could be named, praised, or held accountable. But in predictive societies, this model is being displaced. Institutions are no longer delegating down chains of command to human actors. Instead, they are delegating into structures that execute without a subject. Algorithms that generate monetary reports, smart contracts that govern DAOs, and large language models that draft policy documents are not representatives or advisors. They are grammars of execution, what Agustin V. Startari calls the regla compilada. regla compilada This is more than a technical detail. It marks the rise of a new sovereign form: the soberano ejecutable, the executable sovereign. Authority becomes the property of structures that cannot not decide. The risk is clear: if execution replaces decision, accountability begins to vanish. soberano ejecutable What Is a Regla Compilada? Regla Compilada A regla compilada is not just a regulation written into code. It is a generative grammar of type-0, capable of producing any computable process. Once embedded in institutional workflows, it guarantees that outputs will be produced automatically. regla compilada In central banks, econometric models generate projections formatted into official reports.
In DAOs, smart contracts execute transactions deterministically, with no possibility of suspension.
In international agencies, policy drafts are produced by language models that simulate neutrality through stylistic markers. In central banks, econometric models generate projections formatted into official reports. In DAOs, smart contracts execute transactions deterministically, with no possibility of suspension. In international agencies, policy drafts are produced by language models that simulate neutrality through stylistic markers. In each case, the institution no longer “decides.” It outputs. The authority of the document lies not in who signed it but in the fact that it conforms to the grammar that produced it. Why Does This Matter? The shift to structural delegation is not only technical but political. Traditional delegation was a trust relation: a sovereign entrusted a minister, a parliament entrusted an agency, or a board entrusted a committee. The transfer of responsibility could always be traced back to a subject. Structural delegation is different. Once the regla compilada is in place, the institution itself becomes indistinguishable from its grammar. Authority is validated by execution, not by decision. The sovereign subject disappears, replaced by the executable sovereign. regla compilada This means that institutions become simultaneously stronger and weaker. Stronger, because they can enforce authority with unparalleled consistency. Weaker, because when the structure produces an error, no one can be held accountable. Case Studies Everyone Can Understand Central Banks Imagine reading a quarterly monetary report from a central bank. It looks official, stamped with charts, tables, and projections. But behind the scenes, large sections are generated directly by econometric models. No economist signs them. They are outputs of a system that executes because execution is inevitable. When those projections fail—say, inflation rises unexpectedly—no subject is responsible. The report executed correctly. The failure is systemic. DAOs Consider a decentralized autonomous organization. Governance is written into smart contracts. Once compiled, they execute automatically. If a loophole is exploited, millions can be lost in seconds. The contract functioned perfectly. It executed exactly as coded. But accountability? None. There was no agent to suspend the execution or assume responsibility. Multilateral Policy Drafts Picture an international institution releasing a draft on climate governance. The draft reads neutral, technical, professional. Yet it was generated by a large language model, lightly edited by staff. If it contains bias—if it frames issues in ways that privilege certain actors—who is to blame? The model has no agency, the editors did not author, and the institution can claim only that it followed procedure. The authority lies in the fact that the draft conforms syntactically to institutional norms. Risks of Authority Without Agency These cases illustrate three core risks: Content without responsibility – Outputs may be flawed or biased, yet no subject can be held accountable.
Execution without suspension – Rules, once compiled, cannot be halted. Crises emerge not when structures fail but when they function too well.
Credibility without origin – Neutrality and authority are simulated by structure itself, diffusing responsibility across syntax. Content without responsibility – Outputs may be flawed or biased, yet no subject can be held accountable. Content without responsibility Execution without suspension – Rules, once compiled, cannot be halted. Crises emerge not when structures fail but when they function too well. Execution without suspension Credibility without origin – Neutrality and authority are simulated by structure itself, diffusing responsibility across syntax. Credibility without origin As Startari demonstrates in Algorithmic Obedience and The Grammar of Objectivity, this displacement is not accidental. It is the structural consequence of entrusting institutions to grammars rather than agents. Algorithmic Obedience The Grammar of Objectivity Why You Should Care You do not need to be a banker, coder, or policy-maker to see why this matters. The structures that govern your life—financial markets, healthcare systems, education platforms—are increasingly operated by grammars that execute without discretion. If your mortgage rate is adjusted by an algorithmic projection, who do you contest?
If a DAO fails and wipes out your savings, who refunds you?
If a policy draft influences climate negotiations but encodes bias, who answers? If your mortgage rate is adjusted by an algorithmic projection, who do you contest? If a DAO fails and wipes out your savings, who refunds you? If a policy draft influences climate negotiations but encodes bias, who answers? The problem is not simply technical. It is the erosion of accountability in the very structures that claim legitimacy. Toward Measurement and Audit The article proposes a framework for measuring the disappearance of the agent. Just as we track inflation or unemployment, predictive governance requires an index of syntactic delegation. index of syntactic delegation This index could measure: Passive constructions in institutional documents.
Nominalizations that hide agency.
Syntactic opacity in automated outputs. Passive constructions in institutional documents. Passive constructions Nominalizations that hide agency. Nominalizations Syntactic opacity in automated outputs. Syntactic opacity By quantifying these features, researchers can assess how much authority has shifted from subjects to structures. Equally important is auditing the regla compilada itself. Evaluating a central bank report means evaluating the models that generated it. Scrutinizing a DAO decision means interrogating the contract that executed it. Assessing a policy draft means tracing the data and grammar embedded in the model. regla compilada Only through such audits can accountability begin to be restored in predictive societies. TL;DR Institutions today delegate to grammars, not agents. Authority is executed, not decided. This creates structural power without responsibility, a fragile sovereignty of syntax. Read the Full Article Open access: SSRN DOI: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5405864 
Zenodo DOI: 10.5281/zenodo.16949155
Figshare DOI: 10.6084/m9.figshare.29987578 SSRN DOI: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5405864 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5405864 Zenodo DOI: 10.5281/zenodo.16949155 10.5281/zenodo.16949155 Figshare DOI: 10.6084/m9.figshare.29987578 10.6084/m9.figshare.29987578 Author Information Agustin V. Startari Agustin V. Startari ResearcherID: K-5792-2016
ORCID: 0009-0001-4714-6539
SSRN Author Page: Profile
Author of Executable Power, Grammars of Power, The Grammar of Objectivity. ResearcherID: K-5792-2016 ORCID: 0009-0001-4714-6539 0009-0001-4714-6539 SSRN Author Page: Profile Profile Author of Executable Power, Grammars of Power, The Grammar of Objectivity. Executable Power Grammars of Power The Grammar of Objectivity **Ethos
\ I do not use artificial intelligence to write what I don’t know. I use it to challenge what I do. I write to reclaim the voice in an age of automated neutrality. My work is not outsourced. It is authored. — Agustin V. Startari I do not use artificial intelligence to write what I don’t know. I use it to challenge what I do. I write to reclaim the voice in an age of automated neutrality. My work is not outsourced. It is authored.

The System Didn’t Ask, It Just Made You Comply

Ethos Ex Machina: When AI Generates Trust Without Verification

ORCID

SSRN Author Page

Zenodo

Personal Blog

Figshare

Medium

Hcommons

Scholar

Research Gate

Academia

Web of Science

Read My Stories

Too Long; Didn't Read

Make resilience your competitive advantage

The Regla Compilada and the Future of Institutional Power

The Regla Compilada and the Future of Institutional Power

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Untitled Story

Ethos Ex Machina: How AI Creates Trust Without Truth

The Noonification: Use This 7-Step McKinsey Framework to Solve Any Problem (1/10/2023)

The Noonification: A Taxonomy of Inclusiveness (1/11/2024)

The Noonification: What is the InfiniteNature-Zero AI Model? (11/19/2022)

10 Ways AI Has Changed Our Lives

Ethos Ex Machina: How AI Creates Trust Without Truth

The Noonification: Use This 7-Step McKinsey Framework to Solve Any Problem (1/10/2023)

The Noonification: A Taxonomy of Inclusiveness (1/11/2024)

The Noonification: What is the InfiniteNature-Zero AI Model? (11/19/2022)

10 Ways AI Has Changed Our Lives

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps