LinkedIn·Saturday, 15 August 2026·10d ago

Just finished mapping the 13 most-cited AI governance frameworks against each other. Surprising findings : → Only 2 of 13 offer a formal…

Aleph Technologies

37 followers

Just finished mapping the 13 most-cited AI governance frameworks against each other. Surprising findings : → Only 2 of 13 offer a formal certification or trust label → Only 1 of 13 does formal verification at runtime → Just 5 of 13 were designed agentic-native from the start → Multi-agent governance? Addressed by 4 of 13 The frameworks were not built to be compared. They emerged from different worlds : academic research, vendor strategy, national regulators, open-source communities. Most of them predate agentic AI as the dominant deployment model. So when organizations pick a framework based on vendor familiarity or certification visibility, they're picking on signals that probably don't actually map to what the framework will do for them. This is Article 9 in our 10-part series on EU AI Act governance. It breaks down the 8 dimensions that differentiate frameworks (maturity, scope, agentic specificity, regulatory alignment, runtime governance, certification, multi-agent governance, open-source tooling) and ships with a structured worksheet for evaluating any framework you're considering. Article link in the first comment. #AIGovernance #EUAIAct #AgenticAI #AICompliance #ResponsibleAI

💬 1↻ 1

View on LinkedIn