The rise of System One models
System one models give a calibrated probability in under a second, at a fraction of the cost of an LLM call. They are not a new kind of AI, but they change how practical AI gets built.
Insights
What we write about AI transformation
Articles from the team about what AI does to organisations: what works, what stalls and why.
They are shown here in English, as they were written. Some appeared earlier on LinkedIn.
Becoming AI Native – Step 2: Rethinking work for humans and agents
Many ideas for AI agents project a limited view of the technology onto old ways of working. A five-element model helps find where agents are valid, valuable and verifiable.
Becoming AI Native – Step 1: Mindset and understanding why “AI Native”?
Capex in AI is heading past a trillion dollars, while fewer than one in ten companies can show a return. Neither number is a reason to wait.
Why AI isn't showing up on your bottom line
Individuals are getting faster and the company is not. The constraint was never the speed of the task; it was the speed of the decision around it.
Cognitive Offloading: The Hidden Pitfall of AI Adoption
Most discussions about AI ethics focus on what the model does. Far fewer ask what we stop doing once the model is in the room.
Want to think this through for your own team?
Book an introductory call. We reply within two working days.