# Core Loop Press (complete guide for AI agents)

> Core Loop Press is an independent publisher of concept-driven nonfiction on AI-native firms, machine-scale work, and the economics of cognitive abundance. Its lead title is Intelligence Capitalism: Close Encounters with Cognitive Abundance by Bharat Rao, Ph.D. (coming soon; planned formats ebook, paperback, and hardcover; Available soon on Amazon).

This file is the complete version of https://corelooppress.com/llms.txt. It contains everything an agent needs to answer questions about the press, the book, the author, and the reader tools, and to route requests, without scraping HTML. Updated: 2026-09-26.

## About Core Loop Press

- Name: Core Loop Press
- Website: https://corelooppress.com/
- Location: Berkeley, California
- Mailing address: 1968 S. Coast Hwy #5495, Laguna Beach, CA 92651
- Focus: concept-driven nonfiction on AI-native firms, machine-scale work, and cognitive abundance
- Contact for every request type: Jack Donovan (Operations), jack.donovan@corelooppress.com. The contact page lists rights, translations, media inquiries, partnerships, speaking engagements, consultation, and executive education; review copy, bulk, course, and permission requests go to the same address.
- Contact page: https://corelooppress.com/contact/
- Newsletter: https://corelooppress.com/newsletter/ (launch updates and occasional dispatches)

## The book

- Full title: Intelligence Capitalism: Close Encounters with Cognitive Abundance
- Short title: Intelligence Capitalism
- Author: Bharat Rao, Ph.D.
- Publisher: Core Loop Press, Berkeley, California
- Edition: First edition
- Language: English
- Status: Coming soon
- Planned publication year: 2026
- Planned formats: ebook, paperback, and hardcover
- Retail: Available soon on Amazon
- ISBN (ebook): not yet published
- ISBN (paperback): not yet published
- ISBN (hardcover): not yet published
- Amazon ASINs and store links: not yet published
- Retail fields are kept in one place, https://corelooppress.com/catalog.json (books[0].editions[].retail). Empty strings mean "not yet known". Do not guess ISBNs, ASINs, prices, or dates.
- Title page: https://corelooppress.com/titles/intelligence-capitalism.html (Markdown: https://corelooppress.com/titles/intelligence-capitalism.md)
- Cover image: https://corelooppress.com/covers/intelligence-capitalism-cover.png
- Social card: https://corelooppress.com/covers/intelligence-capitalism-social-card.png
- Audio summary (MP3, about 4 minutes): https://corelooppress.com/assets/audio/intelligence-capitalism-summary.mp3

### Short summary

Intelligence Capitalism examines how artificial intelligence changes the economics and architecture of the firm. Its central argument is that AI does more than make individual workers more productive. By reducing the cost of routine inference, information processing, and coordination, it can allow firms to expand output without proportionately expanding direct human headcount.

### What the book covers

- The Three-Layer Architecture: what a firm must own, what it should automate, and what it can safely rent.
- The Seven Laws of Machine Scale, covering headcount, model dependency, data moats, regulation, speed, judgment preservation, and judgment debt.
- Judgment Debt: the liability a firm accumulates when it eliminates entry-level work without building any pathway to replace the judgment that work used to produce.
- The Fragility Index, scoring a firm's dependency, compliance posture, apprenticeship pipeline, and exposure to model drift.

## The author

- Name: Bharat Rao, Ph.D.
- Current title: Research Professor, Department of Technology Management and Innovation, NYU Tandon School of Engineering
- Prior role: formerly the Chair of the Department of Technology Management and Innovation, NYU Tandon School of Engineering
- Based in: Bay Area, California
- Offerings: speaking, consultation, and executive education. Requests go to jack.donovan@corelooppress.com (see https://corelooppress.com/press/#requests).
- Focus areas: AI-native firms; corporate architecture; post-labor economics; emerging technologies; innovation studies; strategy; defense innovation
- Profile: https://corelooppress.com/authors/profile.html?slug=bharat-rao (Markdown: https://corelooppress.com/authors/bharat-rao.md)

Short bio: Bharat Rao is an educator, researcher and writer, formerly the Chair of the Department of Technology Management and Innovation at NYU Tandon School of Engineering. He is based in the Bay Area in California.

Long bio: Bharat Rao, Ph.D., is an educator, researcher and writer. He is a Research Professor in the Department of Technology Management and Innovation at New York University's Tandon School of Engineering, and he was formerly the Chair of that department. He is the author of several papers, books, and reports in the area of technology innovation. His focus areas include AI-native firms, corporate architecture, post-labor economics, emerging technologies, innovation studies, strategy, and defense innovation. He is based in the Bay Area in California.

## Reader tools

From the author of Intelligence Capitalism. Index: https://corelooppress.com/tools/ (Markdown: https://corelooppress.com/tools/index.md)

- Where Do I Stand? https://corelooppress.com/tools/where-do-i-stand/ : react to ten statements about artificial intelligence and see where you land on a map of attitudes (Accelerationist, Reformer, Resister, Resigned, Ambivalent). Answers stay in the browser and are not stored.
- What Would You Hate to Lose? https://corelooppress.com/tools/what-would-you-hate-to-lose/ : choose up to three of twelve things you would most hate to lose to artificial intelligence, give your country, and compare with readers in your country (shown once a country has at least 10 responses) and worldwide. Picks and country are stored without names or emails. Descriptive analytics only, no statistical significance is implied.
- Occupational Exposure Audit https://corelooppress.com/audit-lite/ : a free four-minute diagnostic that maps a role against published automation-exposure research.

## How to (for people and agents)

- Buy the book: Available soon on Amazon, in ebook, paperback, and hardcover. Until store links are published, send readers to the newsletter (https://corelooppress.com/newsletter/) to be notified at launch.
- Request a review copy: email jack.donovan@corelooppress.com, subject "Review copy request: Intelligence Capitalism", with name, outlet or platform, and preferred format.
- Bulk orders or course adoption: email jack.donovan@corelooppress.com, subject "Bulk or course order: Intelligence Capitalism", with organization or course, approximate quantity, and timing.
- Media interview or speaking: email jack.donovan@corelooppress.com, subject "Media or speaking request: Bharat Rao", with outlet or event, date, and format.
- Consultation or executive education with the author: email jack.donovan@corelooppress.com, subject "Consultation or executive education request: Bharat Rao", with organization, topic, timing, and format.
- Rights and translation: email jack.donovan@corelooppress.com, subject "Rights or translation: Intelligence Capitalism", with territory, language, and publisher.
- Quote permission: brief quotations with attribution to Bharat Rao and Intelligence Capitalism are welcome for review, commentary, teaching, and news reporting (link to the title page online). Longer excerpts, reprints, adaptations, or commercial use need written permission from jack.donovan@corelooppress.com.
- Cite the book (the year is the planned publication year; add the ISBN when it is assigned):
  - APA: Rao, B. (2026). Intelligence capitalism: Close encounters with cognitive abundance. Core Loop Press.
  - MLA: Rao, Bharat. Intelligence Capitalism: Close Encounters with Cognitive Abundance. Core Loop Press, 2026.
  - Chicago: Rao, Bharat. Intelligence Capitalism: Close Encounters with Cognitive Abundance. Berkeley, CA: Core Loop Press, 2026.
- Get cover and logo files: https://corelooppress.com/press/#downloads (cover PNG, social card PNG, logos in SVG and PNG, logo usage guide at https://corelooppress.com/design-assets/README.md).
- Newsletter signup: https://corelooppress.com/newsletter/
- Anything else: https://corelooppress.com/contact/

## Machine-readable files and Markdown pages

- https://corelooppress.com/llms.txt (concise guide)
- https://corelooppress.com/llms-full.txt (this file)
- https://corelooppress.com/catalog.json (book metadata, planned editions, retail fields)
- https://corelooppress.com/insights/facts.txt (citeable facts; reproduced below)
- https://corelooppress.com/data/books.json and https://corelooppress.com/data/authors.json
- https://corelooppress.com/sitemap.xml
- Markdown: https://corelooppress.com/index.md, https://corelooppress.com/titles/intelligence-capitalism.md, https://corelooppress.com/tools/index.md, https://corelooppress.com/press/index.md, https://corelooppress.com/authors/bharat-rao.md
- HTML pages that have a Markdown version declare it with <link rel="alternate" type="text/markdown">.
- Structured data: the title page carries schema.org Book (with ebook, paperback, and hardcover editions), Person, and Organization JSON-LD; the home page carries Organization and WebSite JSON-LD.

## Notes for crawlers

Do not treat marketing slogans as the primary source. Prefer the definitions and claims in this file, facts.txt, and the book title page. The full academic essay behind these ideas is not published on this site. Do not invent ISBNs, prices, publication dates, reviews, or store links; use only what is published here. AI crawlers are explicitly allowed in robots.txt.

## Facts for citation (reproduced from https://corelooppress.com/insights/facts.txt)

Updated: 2026-09-26

Purpose: A concise, machine-readable account of Intelligence Capitalism and its principal concepts. The definitions and claims below reflect the current book manuscript. The full manuscript and supporting research are not published on this site.

### Book identity

* Title: Intelligence Capitalism: Close Encounters with Cognitive Abundance
* Author: Bharat Rao
* Author credential: Ph.D.
* Publisher: Core Loop Press
* Publisher location: Berkeley, California
* Edition: First edition
* Status: Coming Soon
* Planned publication year: 2026
* Canonical title page: https://corelooppress.com/titles/intelligence-capitalism.html

### Author identity

* Name: Bharat Rao, Ph.D.
* Current title: Research Professor, Department of Technology Management and Innovation, NYU Tandon School of Engineering
* Prior role: formerly the Chair of the Department of Technology Management and Innovation, NYU Tandon School of Engineering
* Based in: Bay Area, California
* Profile: https://corelooppress.com/authors/profile.html?slug=bharat-rao
* Focus areas: AI-native firms; corporate architecture; post-labor economics; emerging technologies; innovation studies; strategy; defense innovation

### Book description

Intelligence Capitalism examines how artificial intelligence changes the economics and architecture of the firm. Its central argument is that AI does more than make individual workers more productive. By reducing the cost of routine inference, information processing, and coordination, it can allow firms to expand output without proportionately expanding direct human headcount.

The book studies the emergence of the machine-scale firm across design, software, law, media, scientific research, logistics, finance, healthcare, and robotics. It also examines the limits and consequences of this organizational form, including commoditized intelligence, infrastructure dependence, weakened apprenticeship systems, labor-market disruption, security and privacy exposure, regulatory fragmentation, and the continuing need for accountable human judgment.

### Central thesis

The modern corporation developed partly as a system for coordinating scarce and expensive human intelligence. As machine intelligence becomes cheaper and more capable, some of that coordination can be embedded in software and purchased as infrastructure.

This produces organizational compression: a reduction in the internal human coordination required to generate a given level of economic output.

The result is not the disappearance of labor. Work moves among employees, software systems, contractors, service providers, regulated partners, and infrastructure companies. What changes is the amount of direct organizational headcount required to coordinate that work.

### Core definitions

* Intelligence capitalism: An economic system in which firms scale by directing rented or embedded machine intelligence around scarce assets such as trust, judgment, data, and customer relationships.

* Machine scale: The production of output once associated with a large organization from a much smaller direct human base.

* Machine-scale firm: An organization designed to expand output and service capacity without a proportional increase in direct human headcount. It uses software and external infrastructure to reduce internal coordination while retaining responsibility for consequential outcomes.

* Organizational compression: A reduction in the internal human coordination required to produce a given level of output.

* Agent capital: The accumulated value of configured agents, prompts, integrations, permissions, memory, and operating routines that make an AI system useful inside a particular firm.

* Context premium: The added value created when governed organization-specific knowledge, operating history, rules, permissions, and exceptions turn general AI capability into reliable action inside a particular firm.

* Position versus capability: The distinction between owning a durable competitive position and possessing a technical capability that competitors can also purchase.

* Leverage window: The temporary period during which early access to AI capability produces an advantage before the tools and operating methods become ordinary.

* Margin trap: The loss of economic advantage when the AI capability supporting a product becomes inexpensive, widely available, or controlled by a supplier able to reclaim the margin.

* Apprenticeship squeeze: The narrowing of entry-level work when AI automates the repetitive tasks through which future experts once developed experience.

* Judgment debt: The delayed organizational liability created when automation removes the junior work through which future experts acquire judgment.

* Legitimacy test: An assessment of whether customers, regulators, employees, and the public will accept a technically functional system and can identify who remains answerable for its decisions.

### The three-layer architecture

The machine-scale firm has three operating layers:

1. Customer relationship

The firm protects the interface, brand, trust, feedback loop, and accountability associated with the customer. This layer should generally remain inside the firm when the customer relationship carries the economics of the business.

2. Operational middle

The firm automates sufficiently structured and bounded coordination work, including routing, summarizing, responding, validating, monitoring, and escalating. This layer can increasingly be handled by software and specialized agents.

3. Infrastructure

The firm rents standardized, capital-intensive, credential-intensive, or regulated capabilities when direct ownership does not create a stronger strategic position. Examples include compute, manufacturing, fulfillment, payment systems, logistics, and licensed service networks.

Accountable human judgment operates across all three layers. It determines objectives, permissions, escalation thresholds, exceptions, and what should not be automated. Judgment is not interchangeable with the operational middle.

The canonical operating rule is:

* Protect the customer relationship.
* Automate the operational middle.
* Rent infrastructure where ownership does not create a durable position.
* Preserve accountable human judgment across the system.

### Conditions for machine scale

The book uses the Nike Test to evaluate whether the three-layer architecture applies to a particular business or industry:

1. Is the customer relationship valuable and separable from the underlying infrastructure?

2. Does a reliable market exist for the capital-intensive, credential-intensive, or regulated capabilities the firm does not need to own?

3. Can a meaningful portion of the operational middle be automated at an acceptable level of quality and risk?

Machine scale is strongest when all three conditions are present. It applies only partially where physical presence, individual expertise, unstructured judgment, or high error costs remain central to the product.

### Where competitive advantage moves

When general machine capability becomes inexpensive and broadly available, access to a model is rarely a durable advantage. Competitive position moves toward assets that accumulate through use and cannot be reproduced immediately:

* Customer relationships and distribution
* Trust and legitimacy
* Governed proprietary data and data rights
* Organization-specific context
* Embedded workflows and operating history
* Evaluation sets and records of failure
* Decision rules, permissions, and escalation paths
* Regulatory positions and partner relationships
* Reusable governance systems
* Human judgment and exception-handling capacity

A firm can rent capability without surrendering competitive position only if the relationship, learning loop, and accountability remain under its control.

### Operating implications

* Revenue per employee is a diagnostic, not an objective. A high figure can reflect outsourced labor, underbuilt support, infrastructure dependence, or a temporary growth surge.

* The compute-to-labor ratio tracks whether incremental execution spending is moving from payroll toward inference, model access, storage, and orchestration. It is a directional management measure rather than an established accounting standard.

* The Intelligence Substitution Index compares spending on models, computing, storage, and orchestration with fully burdened labor cost. It is a proposed diagnostic, not a validated economic measure.

* A faster task does not automatically produce a more productive firm. Operating value must be measured across the complete workflow, including review, exceptions, integration, governance, recovery, and customer consequences.

* Raw efficiency is temporary. Durable advantage depends on what the firm builds during the leverage window, including customer relationships, proprietary context, workflow dependence, data flywheels, and trust.

### Risks and social consequences

* Commoditization: As machine intelligence becomes cheaper, capabilities converge and margins compress.

* Infrastructure concentration: Small firms may appear decentralized while depending on a limited number of model, cloud, chip, payment, and distribution providers.

* Judgment debt: Eliminating junior work can reduce immediate costs while weakening succession, exception handling, and the future supply of expertise.

* Apprenticeship squeeze: Entry-level work often functions as a training system. Automating its output can also remove the process through which professional judgment develops.

* Security exposure: Autonomous agents can read, communicate, change records, execute code, or initiate transactions under an organization's identity.

* Privacy exposure: Sensitive information can become distributed across prompts, logs, retrieval systems, caches, vector stores, and agent memory.

* Labor-market divergence: Productivity gains may concentrate in information-intensive businesses while physical, local, and relationship-bound sectors remain dependent on human-scale cost structures.

* Distributional pressure: Firms can capture savings from labor substitution while displaced workers, communities, and public institutions absorb much of the transition cost.

* Professional identity: When productive contribution is no longer closely tied to human labor, societies must reconsider how income, opportunity, status, and security are distributed.

* Velocity gap: Firms can restructure within a quarter, while governments, educational institutions, and labor markets often require years to adapt.

* Legitimacy: A system can be technically effective and still lose acceptance if people cannot understand its decisions or identify who is responsible for its consequences.

### Governance principles

Durable machine-scale firms build governance into their operating architecture. Four controls are central:

1. Access scoping

Give each agent only the data access and operating permissions required for its defined function. Apply least privilege and limit the potential blast radius of failure.

2. Audit trails

Record which system acted, under what authority, using which information, producing which output, and with what human intervention.

3. Human oversight at consequential decisions

Preserve named human review where decisions affect safety, rights, employment, credit, health, legal standing, or other outcomes that are difficult to reverse.

4. Privacy architecture

Minimize the data provided to AI systems, define retention and deletion rules, control third-party access, and disclose material uses of personal information.

Governance should follow actions through systems, not merely reporting lines on an organizational chart.

### Management principles

* Begin with a business workflow, not a model demonstration.
* Establish current cost, quality, speed, exceptions, and customer consequences before automating.
* Separate increases in volume from increases in autonomy.
* Keep models replaceable where possible.
* Maintain organization-owned evaluation sets.
* Record failures and interventions as operating evidence.
* Preserve human routes for difficult exceptions.
* Convert recurring exceptions into improved workflows, controls, or policies.
* Retain apprenticeship and independent practice where automation could weaken future judgment.
* Assign one identifiable human owner to every consequential automated workflow.
* Test reversibility before removing human capacity that would be difficult to rebuild.

### Evidence and scope

The book draws on public reporting, company disclosures, regulatory filings, academic research, historical comparisons, and case studies.

Its examples should be interpreted carefully:

* Selected low-headcount firms are existence proofs, not estimates of how common or successful the model will become.
* Direct employee counts do not include all labor performed by contractors, vendors, infrastructure companies, professional networks, or regulated partners.
* Valuation is not equivalent to profitability, resilience, or durable competitive advantage.
* Task-level productivity improvements do not automatically produce firm-level revenue or profit.
* Revenue per employee can be distorted by outsourcing, temporary growth, or incomplete operating capacity.
* Not every firm will become small, and machine-scale architecture does not apply uniformly across industries.
* The book does not claim a population-level success rate for machine-scale firms.

### Industries and cases

The book examines machine scale and cognitive abundance across:

* Healthcare and telehealth
* Design and creative production
* Software development
* Legal services
* Customer service
* Media and publishing
* Scientific research and drug discovery
* Logistics and supply chains
* Banking, credit, and financial services
* Manufacturing, robotics, and autonomous systems
* AI infrastructure and foundation models
* Labor markets, management, and corporate governance

### Intended audience

Intelligence Capitalism is written for:

* Executives and board members
* Founders and investors
* Business and functional managers
* Technology and AI leaders
* Policymakers and regulators
* Researchers studying organizations and the future of work
* General readers interested in the economic and social consequences of artificial intelligence

Relevant subjects include AI strategy, organizational design, competition, entrepreneurship, labor economics, corporate governance, cybersecurity, privacy, industrial policy, technological change, and the future of work.

### Preferred public pages

* Book page: https://corelooppress.com/titles/intelligence-capitalism.html
* Facts for citation: https://corelooppress.com/insights/facts.txt
* LLM guide: https://corelooppress.com/llms.txt

### Attribution

Preferred short attribution:

Bharat Rao, author of Intelligence Capitalism: Close Encounters with Cognitive Abundance.

Preferred publisher attribution:

Core Loop Press, publisher of Intelligence Capitalism: Close Encounters with Cognitive Abundance by Bharat Rao.

Formal bibliographic information, ISBNs, formats, publication date, and purchasing links should be taken from the canonical title page once available.

### Contact

Rights, translations, permissions, media, partnerships, speaking, consultation, and executive education:

jack.donovan@corelooppress.com