Author · Consultant · Artificial Intelligence
Practical artificial intelligence for legacy enterprises — and for the people who must decide on it without being technical.
About
I bridge what AI actually does and what traditional businesses actually need.
Fabricio Alves Cunha, MSc, is a Brazilian author and consultant working at the intersection of artificial intelligence, technology, and strategy, with a focus on legacy enterprises. His work translates technical AI concepts for non-technical business audiences, always grounded in what works, what it costs, and what changes on the ground.
He writes and publishes in Portuguese and English, and produces regular content on corporate AI adoption, the economics of AI models, and the impact of automation on cognitive labor.
Books
The Challenges and Opportunities of A.I. · Portuguese original: Inteligência Artificial para Empresas Legadas
A practical guide on how traditional businesses can adopt AI without empty promises — from the initial diagnosis to implementation decisions, written with the reality of those who run established systems and processes in mind.
View on Amazon →How to Use AI to Write Books and Scientific Papers Without Sounding Like AI · Portuguese original: A Escrita Invisível
A practical manual on using AI to write books and scientific papers without producing text that sounds artificial — an honest diagnosis, fixes for the typical tells of machine-generated writing, and real craft. Fully adapted for English-language readers.
View on Amazon →Turn your knowledge into income with technology — in Portuguese
A guide for professionals and entrepreneurs aged 45 and over to put AI to work for them — turning accumulated experience and knowledge into value, without being left behind in the technological shift.
View on Amazon →Research
Original title: Agentes de I.A. e a substituição do trabalho cognitivo
A paper examining why the automation of cognitive labor — present since the 1980s — changes in nature from 2022 onward: large language models and agents reach non-routine tasks and affect early-career skilled workers first, read against Brazilian labor-market data.
Read the article →Projects
AI for reading and writing in Brazilian public schools
An AI agent that teaches children to read and write without ever doing the work for them: instead of ready-made answers, it asks Socratic questions at the right moment, so each student writes in their own voice. It runs as 16 structured sessions across 6 pedagogical blocks — one per day, building on each other — with 100% of the text written by the student. Works in any browser, with no installation and an offline mode.
The method follows the same model as a World Bank randomized controlled trial (Edo State, Nigeria, 2025): a 0.31 standard-deviation gain — equivalent to 1.5–2 years of schooling — at US$48 per student. The pedagogy draws on language-education research (Vygotsky, Graham & Perin, Rosenshine), and child privacy is a design constraint: no personal data ever reaches the model and conversations stay on the school network, in line with Brazil's LGPD.
Visit Tutor na Escola →Where to find me
These are the official profiles. They are written identically across LinkedIn, Amazon, ORCID, and Google Scholar — that consistency is what keeps me distinct from other people with the same name.