In Roman mythology, Janus is the god of transitions - guardian of thresholds, beginnings, and ends. He’s famously depicted with two faces: one looking backward to the past, and one forward into the future
That image perfectly captures where many organisations find themselves today when it comes to employee skill development.
Senior professionals in knowledge-based occupations – e.g. lawyers, accountants, management consultants - often reflect on the formative years of their careers: poring over documentation, reconciling data by hand, preparing endless drafts. These weren’t glamorous tasks, but they were seen as essential. Through repetition and exposure, people built their analytical muscle, business instincts, and professional judgement.
But what happens when AI now handles much of that groundwork? What does the future of skill development look like?
It’s a real question facing many professions and, for some, a real worry. If early-career professionals no longer spend hours doing the hard yards, will they still develop the depth of thinking and nuance that senior roles demand?
The concern is valid. But it’s also an invitation.
What’s really at stake?
The anxiety isn’t really about AI; it’s about what gets lost if traditional, foundational work disappears without anything intentional to replace it. And across professions, the challenge looks similar
- In accountancy, AI pre-fills audit trails, but do juniors understand what should trigger concern?
- In consulting, AI drafts strategy options, but can analysts spot the client-specific nuance or stakeholder dynamics?
- In law, AI proposes clauses, but can early-career lawyers weigh their relevance in context?
Repetition used to be the teacher. Now it’s automation. And while that’s a leap forward in efficiency, it creates a new question:
How do we design capability-building when experience is no longer guaranteed through task volume alone?
Reframing the transition in four ways
Here’s how we can reframe this transition - not as a loss, but as a progression. This isn’t about removing thinking, it’s about shifting where and how it happens.
1. From Volume to Value
Instead of throwing junior professionals into mountains of documentation, focus their time on insight, pattern recognition, questioning and interpretation.
AI can summarise a clause, reconcile a ledger, or draft a deck - but the person still needs to ask better questions around:
- What’s missing?
- What’s ambiguous?
- What’s the commercial or reputational risk?
2. From Gatekeeping Tradition to Modernising Capability
Progression has long been built on endurance - those who “did their time” earned the right to lead. But this model favours the privileged, burns out talent, and doesn’t always build better judgement.
AI gives us permission to decouple effort from value, to focus on building real capability, faster and more intentionally.
3. From Learning by Osmosis to Learning Out Loud
Much learning was previously designed to be “on the job” - absorbed through proximity and exposure. But AI changes that, and so does hybrid work. If AI removes opportunities to learn from senior professionals, we must actively replace that visibility.
- Encourage seniors to narrate their judgement aloud
- Debrief real decisions with juniors/newly qualified
- Share examples of when AI got it wrong - and what human insight corrected it
4. From Passive to AI-Enabled Apprenticeships
AI can become a powerful developmental tool - if we design for it.
Instead of assigning juniors only the leftover tasks, create structured learning moments around AI-generated work:
• Junior professionals annotate AI outputs to identify what’s right, wrong, or missing
• Teams run peer-to-peer reviews of AI-assisted work to develop collective judgement
• Supervisors set intentional edge cases-scenarios where AI is likely to struggle to stretch analytical thinking
• Professionals practise ‘explain-back’ methods, defending or revising AI-generated insights in front of seniors
• AI itself can be trained as a mentor, suggesting alternate interpretations or prompting reflection
This apprenticeship model has the potential to be more targeted, transparent, and teachable - equipping people with judgement rather than just task experience. Developed further, it replicates the stretch, reflection, and contextual judgement that traditional paths once built - but in a more focused, scalable way.
There be sacred cows
To move forward, firms must also confront the unspoken beliefs that resist change. For example, many Partners unconsciously equate their own path with the "right" path and “essential rites of passage”. This reflects:
- Status quo bias – assuming “the old way” is the best way
- Survivorship bias – forgetting those who burned out or left
- Hindsight bias – overestimating how well the old model worked
- Effort justification – valuing difficulty over effectiveness
This is where Janus offers perspective. Looking back, we see the traditional path to expertise - long, repetitive, and potentially inefficient. Looking forward, we see the opportunity: smarter job design, richer learning experiences, and faster development of the skills that really matter.
Not through volume. Through design.
A powerful starting point?
Ask leaders to reflect on their own early experiences:
· What truly shaped them?
· What could have been better?
· What would they design differently now?
· What areas are you most concerned about AI eroding?
· What is the potential for AI to create more value in skill development that you wish you could have had?
Final Thought
The challenge isn’t that AI is taking development opportunities away. It’s that the pathways to expertise have long been fragmented, inherited, and hard to navigate. This is our chance to redesign them - with intention, clarity, and purpose.
And like Janus, we can honour what brought us here, even as we build something better ahead.
Contact us to learn more about how we can support your AI adoption & skills journey.
