Practical writing on applied AI
Evidence-led articles from the Applicat AI team on getting AI into production: definitions, methods, model selection, governance and the economics of running a system after go-live. Written for the leaders and engineers who have to make it work.
Market
9 min read
The model companies are now the services companies. What that means for you
In 2026 OpenAI, Anthropic, Microsoft and AWS launched deployment businesses and Accenture bought Faculty. What the labs' move into services means for you.
ReadGovernance
6 min read
Agentic AI governance: a ten-point checklist before an agent goes into production
Ten controls to put in place before an AI agent acts in production: least-privilege access, bounded tools, human checkpoints, evaluation gates and logging.
Engineering
7 min read
Frontier models in the enterprise: how to choose, evaluate and switch without rebuilding
Frontier models change every few months. How to select models by evaluation, build model-agnostic architecture and manage upgrades as routine.
Delivery
8 min read
Why 95% of enterprise AI pilots never reach production, and what the 5% do differently
MIT found 95% of enterprise GenAI pilots deliver no measurable P&L impact. The causes are structural and fixable. What the successful 5% do differently.
Definitions
6 min read
What is applied AI? A practical definition for 2026
Applied AI is the practice of applying existing AI capabilities, especially frontier models, to specific business processes for measurable results.
Bring the frontier into production.
Tell us about the process you want to change. We reply within one business day.