When Intelligence Became Cheap

The series Practicum: tools that turn ideas into decisions and action.

New: the weekly AI Does the Work. People Own the Outcome brief

LIVE BRIEF · 8 TO 10 MINUTES · ONCE A WEEK

AI Does the Work. People Own the Outcome.

Where AI meets reality. Each new issue covers one major, verifiable shift and 10 to 15 signals: what changed for people, business, and trust in action.

PRIMARY SOURCES · FACTS SEPARATED FROM INTERPRETATION · ONE REASONABLE NEXT MOVE

Read the latest issue →Subscribe to the digest →

The principle

We give you direction, tools, and links. You choose how deep to go. Each page has a review date because the field keeps moving. Check that date. The main rule is simple: do not trust it until you check it.

Arrived from the book?

Open the Practicum chapter by chapter, an index from a chapter to the material that continues it. Find your chapter and take one material.

Who are you? Choose a door

Five doors lead to five practical first steps. Each one starts with an action, not a reading list.

New to AI

You have barely used an AI assistant, or never opened one. Learn which one to open first, what to do in your first 15 minutes, and how to stay safe without getting buried in jargon.

First step: choose an assistant and ask it three questions from the book.

Start from zero

Professional

How will AI change your work? Choose your profession from 13 roles. See what is getting cheaper, what is becoming more valuable, what you can automate, and where the red flags are.

First step: find your role and sort one week of work into the three baskets.

Browse the professions

Manager

Work through teams and processes with agents. Decide what AI can handle, what must stay with a person, and whether a process has really changed or only gained an AI wrapper.

First step: run one process through the maturity check.

Volume 2 workbook

Small business owner

Find where AI can pay off in a small business. Use the SMB staff map to compare functions and pick a starting point without wasting money.

First step: find the function with the fastest path to cash impact.

SMB staff map · where to start checklist

Everyday user

Voices, faces, and reviews are now cheap to fake. Set up a family rule in one evening, then make your own actions easier for other people to verify.

First step: agree on a code phrase and a callback rule.

Personal trust perimeter · family trust protocol

Where are you now, and what comes next?

The Practicum is a path, not a library. The maturity route shows the stages, your profile records where you are, and the changelog shows what moved since your last visit.

Maturity route

Five stages: get oriented → understand your work → build a system → work through teams and processes → manage agent systems. Each stage has an artifact and a transition test. There are no promised timelines.

Find your stage

My Profile

See completed tools, your current stage, and one next step. The profile stays in your browser and is never sent anywhere. You can export it to Markdown or reset it.

Open My Profile

What is new

The Practicum keeps changing. The changelog groups updates into waves and preserves the release history. If you return after a month, start there.

Open the changelog

How the Practicum works

It has three axes. Volume 1 focuses on professions: you examine your own work. Volume 2, Business in the Age of AI Agents, focuses on business functions: a manager examines processes. Book 3, The Economics of Trust, focuses on verifiability: anyone can ask what to trust and how to prove their own claims. Volume 3 is deliberately smaller. It does not create one page per chapter. Its tools live in three homes: protection, a verifiable trail, and a compass.

Every tool follows the same cycle: diagnosis → artifact → action → check. The volume workbooks above give you an end-to-end route. The method map puts the whole system on one page.

Method map · all tools · worked examples

04 / About the author

Industrial scale

Denis Ermilov

Director and Head of FinTech & AI Practice. Denis began his career at Exigen Services, now Emergn, an international firm that at the time ranked among the world's four leading outsourcing companies. He worked in an English-speaking distributed team with US clients on technology projects for investment businesses in the United States and Europe.

He later worked on central bank payment infrastructure, banking platforms, and international payments: cross-border systems, ISO 20022, China through UnionPay and CIPS, and India through NPCI and RuPay. His project experience includes VTB, Gazprom, McKinsey & Company, Volkswagen Group AG, and Alibaba/Alipay. His open AI-native systems are available on GitHub.

ExigenTop-four global outsourcingUS clients · investment technology
Central bankPayment infrastructure · CBDCNational-scale systems
China · IndiaUnionPay · CIPS · NPCI · RuPayCross-border · ISO 20022
GitHubPublic AI-native systemsOpen source · production
When Intelligence Became Cheap