Mario Thomas

From Hype to Accountability: The Agentic AI Reckoning

Gartner's prediction that over 40% of agentic AI projects will be cancelled by 2027 reveals a sobering truth: the gap between AI enthusiasm and AI governance has never been wider. This month's articles chart the path from understanding what agentic AI actually is to establishing the accountability frameworks that make it sustainable.

I begin by stripping away the hype to reveal agentic AI's core truth: it's generative AI in a loop, where machines drive iteration instead of humans. From there, I expand the definition to compound loops that coordinate multiple AI disciplines simultaneously—machine learning, computer vision, and NLP working together to create exponential rather than linear value. The accountability gap article confronts an uncomfortable reality: whilst organisations race to delegate decisions to AI, the responsibility for those decisions remains firmly with humans who often lack the verification capability to ensure quality.

World models point to AI's next frontier—systems that don't just analyse what has happened but simulate what will happen next. And minimum lovable governance provides the operating principle that makes all of this work: not the smallest amount of governance you can get away with, but the smallest governance system that achieves necessary guardrails whilst being something people actually embrace.

If your time is limited, I particularly recommend the accountability gap piece. The pattern of AI-generated errors reaching courts, clients, and regulators is accelerating, and the strategic choice between augmentation and replacement carries consequences that many Boards haven't fully considered.

How is your organisation balancing the speed of AI adoption with the governance maturity needed to sustain it?

-Mario

Agentic AI: Strip Away the Hype and Understand the Real Strategic Choice

Published 2 November 2025 | 8 minute read

Agentic AI has become this year's poster child, dethroning generative AI as the technology everyone wants to discuss. This article demystifies the hype by revealing the core truth: agentic AI is generative AI in a loop, where the machine drives iteration instead of a human. The strategic question isn't about technology sophistication but where to consciously transfer decision-making agency from people to systems.

The Compound Loop: Why Agentic AI's Real Power Lies Beyond Generative AI

Published 9 November 2025 | 9 minute read

McKinsey's 2025 research shows whilst 88% of organisations use AI, only 23% have successfully scaled agentic systems—and even fewer integrate disciplines beyond generative. This article expands the agentic AI definition to compound loops that coordinate multiple AI disciplines simultaneously, creating interaction effects that multiply capabilities and enable Boards to tackle broader business challenges.

The Accountability Gap: When AI Delegation Meets Human Responsibility

Published 16 November 2025 | 10 minute read

While organisations transfer decision-making agency to AI systems, accountability remains with humans—yet Boards approve AI deployment without investing in the verification capability needed to ensure it. This article demonstrates why augmentation preserves expertise pipelines whilst replacement destroys capabilities that cannot be rebuilt, turning apparent cost reduction into systematic competitive disadvantage.

World Models: The Next Horizon in AI for Predictive Enterprise Intelligence

Published 23 November 2025 | 9 minute read

World models mark AI's shift toward true predictive power, allowing systems to simulate future scenarios and help businesses move from reacting to events to anticipating them. Drawing on emerging research including Yann LeCun's work on simulation-based intelligence, this article highlights the practical gains industries are seeing and the opportunity for Boards to turn future insight into present advantage.

Minimum Lovable Governance: The AI Operating Principle Boards Should Use

Published 30 November 2025 | 8 minute read

Minimum lovable governance marks a shift from episodic compliance scrambles to continuous, embedded oversight that people actually want to use. This article explains how governance can achieve necessary guardrails whilst earning adoption rather than resistance—the operating principle that makes AI governance work when traditional approaches simply get routed around.

Copyright © 2025 Mario Thomas. All rights reserved.

You are receiving this email because you signed up to receive it on mariothomas.com on {{date_subscribed}} at {{time_subscribed}}. You can unsubscribe at any time.