Notes from the decision layer.
Perspectives on decision intelligence, AI agents, and the shift from dashboards to decisions — from the team at Diwo.
I approved 26,043 lines of code in two minutes
We ran Provenance on our own repositories. It found a 26,043-line AI-written pull request approved in two minutes, with no comments. The approver was me.
Slopsquatting: The Supply-Chain Attack Your AI Agent Invites In
Slopsquatting is a supply-chain attack where attackers register the fake package names AI coding agents hallucinate. Why…
AI Code Governance: A 2026 Field Guide
AI writes a growing share of your code. AI code governance is how you prove it's controlled — attribute it, classify it,…
The EU AI Act After the Digital Omnibus: How to Be Evidence-Ready
The EU’s Digital Omnibus moved the AI Act’s main high-risk deadlines to 2027 and 2028. The evidence questions didn’t mov…
How to Govern AI-Written Code Without Buying Another Scanner
You already run 4–8 scanners and carry 50,000 open findings. Governance isn't more detection — it's the decision layer a…
Your AI Agent Has Your Production Keys — Here's How to Find Out
An AI agent running in auto-approve mode on a machine with production credentials approves its own actions, with no huma…
Who's Accountable When AI Writes the Code?
When AI-authored code ships and something breaks, whose name is on the change? Accountability doesn't transfer to a tool…
141,006 Sessions Later: Why AI Incidents Get Excavated, Not Decided
Ungoverned when the code was written, unrecorded once the agent was live, nobody decided. Why the frontier labs' August …
Power BI Alternative: Why Enterprises Choose Diwo for Decisions
Looking for a Power BI alternative? See why enterprises use Diwo for quantified recommendations, audit trails, and decis…