Read this for your industry
Optional. Choose your industry and this piece opens with a situation from your world and closes with a first move for you. Your choice carries across the guides and analysis on this site.
You lead a licensee with a thin adviser pipeline
Adviser numbers fell from almost 29,000 at the start of 2019 to 15,127 in March 2026, and only 464 of today's advisers entered the profession in 2025. The FAAA is pressing for education reforms to make entry easier.
Paraplanning and associate roles have long been where advisers learnt the craft. If agents take on that preparation work, licensees need another way to build judgement in new advisers.
What this means for you
- Keep a structured path for new advisers that includes reviewing agent work with a senior adviser.
- Use agents to show new advisers how experienced advisers reason through a case.
- Recruit from adjacent professions, such as accounting, and use agents to shorten the learning curve.
You lead a territory where the path to store manager is narrowing
Stanford research found substantial employment declines for early-career workers in the occupations most exposed to AI, including customer service. Many retail leaders started on the shop floor and learnt the job by doing the routine work.
As agents handle more routine customer service and administration, area managers need to make sure new staff still get the experiences that build judgement.
What this means for you
- Keep entry-level roles that teach the floor, and give new staff time on the work agents cannot do.
- Use agents to coach new staff in the moment, drawing on what your best managers know.
- Identify future store managers early, and plan the experiences they need.
You lead a workforce that needs 17,000 more carers a year
CEDA estimates aged care needs at least 17,000 more direct care workers each year over the next decade just to meet basic standards of care. The workforce is also ageing, so experienced staff are retiring as demand grows.
Experience in care is built at the bedside, alongside someone who knows. Agents can help pass that knowledge on, provided leaders keep new staff close to experienced ones.
What this means for you
- Use agents to capture how experienced nurses and carers handle difficult situations.
- Take paperwork off experienced staff so they have time to mentor.
- Recruit from adjacent sectors, and use agents to shorten induction.
You lead engineering while junior roles disappear
Stanford research found that employment of software developers aged 22 to 25 had fallen about 20% from its late 2022 peak, while employment of older developers held up or grew.
Today's senior engineers learnt their craft through the tasks agents now handle. If teams stop hiring early-career engineers, there will be fewer people in ten years who can direct and check the agents.
What this means for you
- Keep hiring early-career engineers, and give them review and design work alongside agents.
- Use agents to explain the codebase and past decisions to new starters.
- Hire experienced people from adjacent industries, and use agents to speed up their induction.
You need electricians the country has not trained yet
Jobs and Skills Australia estimates the country will need about 32,000 more electricians by 2030. Experience in the trades is built through apprenticeships, where people learn by watching and doing under supervision.
Agents can take paperwork off supervisors so they have more time to teach, and they can help apprentices understand the reasons behind each step.
What this means for you
- Use agents for compliance paperwork, and give the time back to supervising apprentices.
- Record how your best electricians handle difficult sites, and let apprentices ask the agent about it.
- Keep apprentices on the hard jobs with experienced electricians.
The junior pathway
Every football club knows its senior side was built in the under-19s years earlier. Cut the junior program and nothing happens straight away. The team keeps winning for a season or two. Then the senior players retire and nobody is ready to replace them.
I have spent my career on the practitioner side of transformation, in retail operations, ecommerce, product engineering and operating models. From where I sit, professional services is cutting its junior program at a difficult time.
AI raises the value of experience
AI now does a lot of what junior people used to do: the first pass of analysis, the research summary, the draft slides. That is a real change, and firms are right to use it.
AI does that work well when someone experienced directs and checks it. That person can look at a demand forecast and see at once that it is nonsense, or spot that the constraint in a replenishment process is the delivery window. That judgement comes from years of watching things go right and wrong.
So AI is raising the value of experience at the same moment the industry is cutting the roles that create it.
The pipeline is narrowing
The evidence is fairly clear. Stanford’s Canaries in the Coal Mine paper, using ADP payroll data, found that workers aged 22 to 25 in AI-exposed occupations saw a 16% relative decline in employment after controlling for firm-level shocks, while employment for experienced workers stayed stable.
Professional services is part of that picture. Business Insider reported that PwC plans to cut US graduate hiring by a third over three years. In the UK, Accountancy Age found the Big Four had collectively scaled back graduate intake, with KPMG’s down 29%.
Firms are making a reasonable short-term call. Graduate programs are expensive, and AI can now do part of what graduates did. The cost appears later, when the firm needs senior people who would have learnt their judgement on the way up.
Experience built in industry
The old consulting path was slow, and it worked. Several years of analytical work under supervision, exposure to different industries, and gradually the judgement that made you useful. It was an apprenticeship, much like a trade.
Some of that experience will now be built in industry. People who spend years running retail operations, clinical services or finance processes will move into advisory roles. In my experience they often become good consultants quickly, because the pattern recognition is already there and they add method on top of it.
In football terms, this is recruiting experienced players from other leagues. It works, and it moves talent between clubs. The total pool of experienced players stays the same size.
Using AI to spread experience
The more promising response is to use agentic tools to make each experienced person’s judgement go further.
Take someone with fifteen years running supply chains. With good tooling, their judgement can be written into briefs, checks and agents that a whole team uses. It can be applied across several organisations at once and passed to a new team in weeks. That person becomes a multiplier for everyone around them.
Experience still has to come from somewhere, and the junior pathway still matters. Agentic tools mean fewer people need to complete the full ten-year path to have the same effect for a client.
The fluency gap with clients
There is one more layer I see often. Advisory firms and technology partners are building agentic capability faster than many of their clients. When a team with high AI fluency works with a team that is still learning, both slow down. The advisers cannot work at the pace their methods allow, and the client cannot fully judge what is being built or why.
The organisations closing that gap fastest treat AI fluency as a capability for the whole workforce and build it broadly across teams. There is more on this in AI fluency is a leadership capability now.
What follows
The talent question in professional services is mostly about where experienced operators will come from as the old pipeline narrows. Job losses are part of the story. A shortage of judgement is the larger long-term risk.
The most valuable profile right now combines deep domain experience with real AI fluency. It is rare, and demand for it is growing.
The gains that compound will come from using agents to pass on what experienced people know. Automating existing tasks saves money. Spreading judgement builds a stronger team, the way a well-run junior pathway keeps a club competitive for decades.
Pair each new adviser with a senior adviser for a weekly review of agent-prepared files, and record what the senior notices that the new adviser missed.
List the experiences that made your best store managers good at the job. Check whether new staff still get each one.
Ask your most experienced nurse to describe five situations every new carer should be ready for. Turn them into short guides an agent can share at the start of each shift.
Look at your hiring plan for the next year. If it has no early-career roles, add one and design what that person will learn in their first six months.
Pick one apprentice to start with. For a month, have an agent prepare their paperwork, and use the time saved for one extra supervised job each week.
Questions readers ask
Is AI taking junior consulting jobs?
The evidence points to fewer entry-level roles in AI-exposed work. Stanford researchers found a 16% relative decline in employment for 22 to 25 year olds in the most exposed occupations, while employment for experienced workers held steady.
What profile is most in demand?
People who combine deep experience in an industry with real AI fluency. That combination is rare, and demand for it is growing.
What should firms do about the pipeline?
Hire more experienced people from industry, and use agentic tools to capture and spread what senior people know, so fewer people need the full ten-year path to have the same impact.
Sources
- Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence (November 2025 revision), Stanford Digital Economy Lab (Brynjolfsson, Chandar, Chen), November 2025
- 4 key ways AI changed the Big Four in 2025, Business Insider, 31 December 2025
- The Big Four's new favourite grad is AI, Accountancy Age, 23 June 2025
The views here are my own, not those of my employer or its clients.