Organisations
Advisory and consulting
AI adoption strategy, automation governance, analytics, and technology operating models. For when the decisions are consequential and the cost of getting them wrong is high.
The most consequential question in AI adoption is not which tools to use.
It is deciding which parts of your organisation to automate, which to govern carefully, and which to keep human. Then building the capability to execute on that decision.
- 15+ yearstechnology leadership, including AMP and CBA
- $10M+technology transformation programmes delivered
- 3 Microsoftcertifications: AI Engineer, Data Scientist, Fabric
- Nowadvising an Australian telco on AI and capacity expansion
The harder question
If AI could run the whole operation, what would you keep human?
The automation question is largely answered. Technical capability is not the constraint. The tools exist and they have never been more accessible.
What is not answered is the more influential side of the decision: which parts you would choose to keep uniquely human, and why.
That is not a hypothetical. It is a strategic design question, and it is worth sitting with before the next tool decision.
Work with me
Different entry points. One approach.
Organisations
Fractional technology leadership
Executive-level technology leadership brought in to provide direction, drive delivery, and leave the organisation stronger.
Organisations
Speaking
Practical talks and briefings for leadership teams navigating adoption, governance, and what it means to be more human with AI, not less.
Build with you · 5 days
5-Day Claude Cowork Sprint
Five guided 1:1 sessions. Daily. Thirty minutes each. Put AI to work without wading through developer tools and acronym frameworks.
Build with you · 5 hours
Claude Cowork Workshop
A live, hands-on online build. Leave with a working AI operating foundation and the ability to keep it compounding.
Build with you · 90 days
Claude Cowork System
Weekly 1:1 sessions to embed an AI operating system calibrated to your context, tools, and ways of working. Built with you, not for you.
The HumanAI scorecard
Are you operating by-design, or by-default?
Fifteen questions, about four minutes. Find your scope of AI adoption, your five domain scores, and the specific constraint on your next step.
Musings
Recent thinking
Leadership · 30 April 2026
Let AI handle the bits so humans can live in the atoms
AI is increasing burnout for many workers, not reducing it. Organisations are using it to do more of the same, not to do something different. The shift required is from chatting to delegating.
AI-assisted work · 17 March 2026
Backprop for humans: how feedback turns AI from generic tool to operational memory
Most people use AI as a disposable answer machine. The organisations that compound their advantage treat it as a long-horizon learning partner. And they train it deliberately.
AI adoption · 17 February 2026
Drawing the bow: why your AI journey must feel worse before it feels better
The J-curve of AI adoption is real. If it does not feel like a step backwards for a while, you are probably not doing real transformation. You are bolting AI onto existing mess.
About
Peter Johnson
More than 25 years in technology, including 15 in senior leadership at AMP and CBA. Programmes valued at $10M+, Microsoft certified across AI engineering, data science, and analytics, and hands-on with AI and machine learning since well before the current wave.
