Use AI to strengthen learning, evidence and professional judgement—not merely accelerate output
I help education providers and institutions decide where AI can genuinely improve learning and work, what people must still understand and judge for themselves, what evidence remains trustworthy, and how policy and implementation should support responsible practice.
“Graeme combines strategic thinking with practical delivery” and has “long been an innovator in the use of technology to improve education and learning.”
— Dr Damon Whitten
The problem institutions are facing
AI can improve visible output faster than institutions can update the systems used to interpret that output. A polished answer may reflect strong learning, weak learning or extensive machine assistance. Tool access alone does not tell an institution what people can understand, judge, adapt or take responsibility for.
The task is therefore larger than adoption. Institutions need to redesign learning, assessment, professional practice and governance so that AI use strengthens capability rather than concealing or displacing it.
When this is useful
This work is useful when:
- staff have AI access but no coherent account of useful, responsible practice;
- assessment-integrity measures are not restoring confidence in what learners understand or can do;
- educators need practical ways to keep consequential thinking visible;
- leaders need policy and governance that can guide decisions without freezing practice;
- professional learning has focused on tools without improving workflows, judgement or evidence;
- an institution wants to prototype AI-enabled practice before committing to a larger system.
Areas of work
AI in education and assessment redesign
Examine how AI changes task design, evidence, authorship, feedback, moderation and confidence in learning.
Visible-thinking and evidence design
Design selected moments where attempts, questions, checks, judgement, application and response to feedback become available for learning and professional interpretation.
Educator and professional capability
Build the practical judgement required to use AI productively, recognise its limits and adapt practice within the relevant domain.
Guided professional conversations
Use structured, responsive dialogue to support reflection, surface judgement and strengthen professional learning without treating conversation as proof on its own.
Policy and governance
Connect principles, permissions, risks, responsibilities and review arrangements to the practice the institution is actually trying to enable.
Prototypes and implementation support
Develop and test bounded workflows, prompts, agents, resources or learning-system components before wider implementation.
What this work produces
The output may include:
- an AI capability and use-case map;
- an assessment or learning-assurance review;
- visible-thinking and evidence-design patterns;
- a professional-learning architecture;
- policy, governance and decision guidance;
- a bounded prototype or guided workflow;
- an implementation sequence with roles, safeguards and review points; or
- an evidence-based account of what should not yet be automated.
The enduring output is a governed system in which AI supports better learning, stronger decisions and durable institutional capability.
Current public proof: Visible Thinking Designer
Visible Thinking Designer is a self-owned, working public prototype. It helps educators redesign activities so that consequential thinking remains visible when polished answers are easy to generate.
Its premise is simple: a finished answer can no longer carry the full evidential burden once placed upon it. Better design makes selected attempts, questions, checks, judgement, application and response to feedback available for learning and professional interpretation—without turning education into a surveillance system.
Read When Answers Become Abundant, Learning Evidence Must Change →
How the work begins
The first commission is usually an assessment, capability or implementation review—not a generic AI workshop.
We identify:
- the decision or practice that needs to change;
- the people and capabilities involved;
- current AI access, use and constraints;
- the evidence, risks and governance requirements; and
- the smallest prototype, redesign or learning intervention worth testing.
Workshops may support implementation once the capability and design question is clear.
Technology and governance boundaries
I am technology-aware and vendor-independent. I help organisations clarify use cases, capability requirements, learning, evidence, policy and implementation.
This work does not include Microsoft 365 tenant administration, product configuration, managed IT services or systems integration. Technical deployment should involve the appropriate specialists and partners.
AI-enabled does not mean AI-led. Human judgement, consent and responsibility remain operative.
Explore related areas of work
If the institution first needs to clarify the capability and evidence problem, explore capability discovery and assurance. If the issue concerns formal evidence, assessment, RPL or recognition, explore credentials, assessment and recognition.
Explore capability discovery and assurance →
Explore credentials, assessment and recognition →
Start with the practice that needs to change
Bring the decision, workflow, assessment or capability question in front of you. We can identify the smallest governed piece of work that will create useful evidence and movement.