Governance Training
Your agents are writing code. Who can say what they did?
The problem the show keeps coming back to
Guests on this podcast come from industry, academia, and the MIT community, and they keep arriving at the same place from different directions: AI is being adopted faster than anyone is governing it.
In software teams that now has a specific shape. Agents open pull requests faster than people can read them, and approvals get given on the strength of a summary. A prompt is a suggestion. A policy is a suggestion someone can waive under deadline.
What the host teaches
Dan Wilson hosts this podcast and founded Info Science AI, where he teaches a governance framework for teams running AI coding agents. It rests on three ideas:
Nothing verifies its own action, and the structure enforces that.
The system names exactly what it cannot catch.
Decisions are recorded where no single party controls them.
The result is a pipeline in which an agent cannot grade its own work, move the bar, narrate its way past verification, or rewrite the record. The framework was built in production and is taught from practice.
Who it is for
- Engineering teams already running coding agents in their pipeline.
- Leads who own the outcome and need an audit trail that survives review.
- Teams that have had an incident and want to know what to change.
How it starts
Every engagement starts with a 15-minute discovery call, at no charge. Sessions are live and remote.
The offerings, the prices, and the form to request a call are on the Info Science AI site.
Governance for agent engineering teams. Built in production. Taught from practice.
Visit Info Science AIThere are very few consultants who know how to do this. You can count them on one hand. If people need consulting in AI assurance or AI governance, go to Daniel.
Greg Hutchins said that on episode 25, a conversation about ISO 42001 and the NIST AI Risk Management Framework. Read the write-up.