Human-Centred AI
Critical thinking
in the age of AI.
Using AI without outsourcing your judgement.
A practical three-hour workshop built around how your team uses AI. Practise checking what the evidence supports, deciding what needs a closer look and explaining your judgement.
The firm recorded fewer complaints per 1,000 customers than the sector average.
The lower complaint rate shows that customers are getting better outcomes. Defer the planned review of customer outcomes.
What would we check before deciding the review can wait?
Illustrative example. Your workshop is built around a decision your team faces.
The workplace challenge
When is it worth pausing a decision?
An AI-assisted recommendation can look ready to act on. Then someone spots something that could change the decision.
When the deadline is close or a senior colleague has already signed off, how do you decide what needs checking and raise it with the team?
-
Missing context
What the summary didn't have in front of it.
-
An overlooked exception
The case that doesn't fit the pattern.
-
An unsupported conclusion
More than the evidence can support.
What your team will practise
Review the evidence. Explain your judgement.
Evaluate an AI output
Separate evidence from assumptions and check an important claim against its source.
“The briefing says our complaint rate is below the sector average. Are both figures calculated the same way?”
Test your own thinking
Form an initial view, then ask AI for an alternative explanation. Check whether the evidence supports it.
“Ask AI for another explanation for the lower complaint rate. What evidence would help us test it?”
Make and explain a decision
Say what you would do, what remains uncertain and what would change your view.
“I'd keep the review on the schedule until we've checked how complaints are recorded.”
Challenge constructively
Practise raising a concern under pressure and discuss how your team could handle it in everyday work.
“Before we sign this off, can we check whether customers find it easy to raise a complaint?”
How we build the pilot
Built around a piece of work your team recognises.
Understand your context
We speak with you about where AI is already being used, what's working well and where errors or uncertainty are showing up.
Design a realistic simulation
Together, we choose one workflow and create a scenario that reflects the decisions your people face. A walkthrough, blank template or redacted example is enough to get started.
Practise and apply
We facilitate the simulation, then help the team identify a review practice it can try in its own work.
Your team uses AI to draft a monthly risk summary for a management meeting.
The AI summary says all actions are on track. The underlying log shows that one action missed its original deadline and is now on track against a revised date.
Your team decides how to report the change and practises explaining that judgement to the manager signing off the report.
Discuss a pilot
Where would better judgement make a difference in your team?
Holistic Training brings over ten years of experience delivering psychologically informed training in communication, leadership and team development. This pilot applies that experience to how people think, decide and speak up when AI is part of their work.
Tell us about one piece of work where your team already uses AI. We'll explore whether it's a good fit for a pilot.