Michelle’s perspective
AI agents
The AI race has a governance problem
The Hugging Face incident exposed something bigger than a model-control problem: the gap between the speed of AI and the capacity of the institutions meant to govern it.
ReadWriting
Writing and briefings.
What I am seeing across AI, organisations and the work ahead.
From Michelle
Michelle’s perspective
AI innovation
The next phase of innovation is not faster outputs
AI is changing how innovation moves from problem to prototype. As outputs become faster and cheaper, the real leverage shifts upstream into problem definition, workflow design, and system thinking.
organisational intelligence
Intelligence is abundant. Workflows aren't ready.
AI is producing intelligence faster than organisations can turn it into workflows, decisions, and services that actually work. The bottleneck has shifted — and the real premium is moving elsewhere.
AI workflows
AI Is Not Just Changing the Tools, It Is Changing the Structure of Work
Many people are already using AI tools, automations, and agents. The deeper opportunity is learning where they truly help, how workflows improve over time, and what it takes to make that work more useful, reliable, and repeatable.
marketing
Marketing feels noisier, judgment matters more
AI has made marketing faster, louder, and more crowded, but not necessarily clearer. The real question now is not whether AI matters, but where it is useful, where it creates risk, and where human judgment still makes the difference.
agentic workflows
Agentic Work Needs Better Rooms, Not More Hype
Last week's Agentic meetup surfaced a clearer picture of how builder work is changing. The strongest signal was not any one tool, but the shift in where constraints now sit, from writing code to framing tasks, reviewing outputs, and shipping safely.
AI strategy
The Real Shift Is Not the Tools, It Is the Structure Beneath Them
A lot of attention is going to AI tools, what matters, what is noise, and what to learn next. That matters, but the deeper shift is structural. The biggest opportunity now sits with networks that build capability together, through shared learning, coordination, and trust.
Signals and observations
What I am seeing
multimodal ai
What Is Multimodal AI, and Why Does It Matter?
A reflection on 25 years from search to multimodal AI, and why organised information, workflows and data structures matter more than ever.
AI
What Happens When AI Stops Being Cheap?
AI has been cheap, abundant, and easy to build upon — but that era won’t last forever. What happens when AI stops being cheap, and what does it mean for builders, appliers, and the next economy?
Interviews, articles and appearances
In the media
Stay connected
Join The Signal.
Get the Weekly Briefing, new perspectives and invitations from The Remix and The Signal Studio.