“What should we budget for AI upskilling?”

This question lands on my desk regularly. It’s a reasonable question without a simple answer. Budgets depend on so many variables: organisation size, existing capabilities, ambition level, internal resources.

But after helping numerous organisations build AI upskilling programs, I can offer frameworks and benchmarks that help with planning.

The Cost Categories

AI upskilling budgets typically include these categories:

Content Development or Acquisition

Getting the learning content itself:

• Internal development: Staff time to create content, potentially months of effort
• External courses: Per-learner license fees for platforms like LinkedIn Learning, Coursera, or vendor-specific training
• Custom development: Engaging consultants to build organisation-specific content
• Hybrid approach: Combining external foundations with internal customisation

Cost range: From near-zero (curating free resources) to hundreds of thousands (comprehensive custom development).

Delivery Infrastructure

How learning reaches people:

• Learning management systems: Platform costs, which may already exist
• Video conferencing tools: For live virtual sessions
• Practice environments: Sandbox tools for hands-on learning
• Physical spaces: Room costs for in-person sessions

Many organisations already have this infrastructure; others need to invest.

Facilitation and Support

People who make learning happen:

• Internal facilitators: Time allocation and potential upskilling
• External trainers: Day rates or program fees
• Coaches and mentors: Time for individual support
• Help desk support: Ongoing assistance for learners

The more human support you provide, the higher this category.

Participant Time

Often the largest real cost, frequently overlooked:

• Training time: Hours employees spend learning rather than producing
• Practice time: Additional hours applying and experimenting
• Manager time: Supporting team development

Calculate this even if you don’t budget it separately—it’s a real resource commitment.

Tools and Technology

AI tools for learners to practice with:

• AI platform licenses: Access to ChatGPT, Microsoft Copilot, or other tools
• Specialised tools: Function-specific AI applications
• Development environments: For technical roles

Some organisations have these already; others need to invest.

Program Management

Running the program:

• Project management: Coordinating activities
• Communications: Promoting and supporting the program
• Administration: Logistics and tracking
• Evaluation: Measuring outcomes

This overhead is often underestimated.

Budget Benchmarks

While every organisation differs, these benchmarks provide starting points:

Per-Person Investment

For general workforce AI literacy:

• Minimal approach: $100-300 per person (curated external content, limited facilitation)
• Moderate approach: $300-800 per person (structured program, some live facilitation, follow-up support)
• Comprehensive approach: $800-2,000+ per person (custom content, extensive facilitation, ongoing coaching)

These exclude participant time costs.

Percentage of Training Budget

Organisations serious about AI capability are allocating:

• 10-20% of training budgets for AI-specific development
• Some are creating separate AI upskilling budgets outside traditional L&D

These benchmarks align with recent AHRI research on workforce development spending trends in Australian organisations.

Phased Investment

First-year programs often require larger investment:

• Year 1: Foundation building, content development, infrastructure
• Years 2+: Maintenance, updates, new topics at lower per-person cost

Plan for ongoing investment, not just initial outlay.

Building Your Budget

A systematic approach to budget development:

Step 1: Define Scope

Who needs what level of AI capability?

• Number of learners at each level
• Depth of capability required
• Timeline for development

Clear scope enables realistic budgeting.

Step 2: Choose Delivery Model

How will you deliver learning?

• Heavy on external content or internal development?
• More asynchronous or more facilitated?
• Centralised or distributed delivery?

Delivery choices drive cost structures.

Step 3: Estimate Each Category

Work through each cost category:

• Get quotes for external content and training
• Estimate internal time requirements
• Calculate infrastructure needs
• Factor management overhead

Be thorough—surprises usually add cost, not reduce it.

Step 4: Calculate Participant Time

Don’t ignore this real cost:

• Hours per learner for formal training
• Additional practice hours
• Multiply by fully loaded hourly cost

This number often surprises stakeholders.

Step 5: Build Contingency

Things cost more than expected:

• 15-25% contingency is prudent
• More for new programs with higher uncertainty

Better to budget contingency upfront than request additional funds later.

Step 6: Phase and Prioritise

If the total exceeds available resources:

• Phase over multiple years
• Prioritise highest-impact populations first
• Start with pilots before full rollout
• Find efficiency opportunities

Realistic phasing is better than underfunded ambition.

Cost Optimisation Strategies

Maximise impact within your budget:

Leverage Existing Resources

• Use current LMS rather than acquiring new platforms
• Adapt existing facilitation capabilities
• Build on current manager development infrastructure

New investment should add capability, not duplicate what exists.

Curate Rather Than Create

• Select from abundant free resources
• Supplement with organisation-specific context
• Focus creation efforts on unique needs

Original content development is expensive. Use it strategically.

Train the Trainers

• Develop internal facilitation capability
• Reduce ongoing reliance on external trainers
• Build sustainable internal capacity

Higher upfront investment for lower ongoing costs.

Scale Strategically

• Use asynchronous content for broad reach
• Reserve intensive facilitation for priority populations
• Layer support by need level

Not everyone needs the most expensive development approach.

Measure and Improve

• Track what works and what doesn’t
• Redirect resources from ineffective approaches
• Continuous improvement reduces waste

Investment in measurement pays returns through efficiency.

Making the Business Case

Budget requests require justification. Frame the case effectively:

Quantify Productivity Gains

If AI-proficient workers are 20% more productive on certain tasks:

• Calculate hours saved across the workforce
• Translate to dollar value
• Compare to training investment

Productivity math often supports significant investment.

Consider Competitive Risk

What happens if competitors develop AI capability and you don’t?

• Market share risk
• Talent attraction disadvantage
• Innovation gap

Risk avoidance arguments complement ROI arguments.

Factor Employee Expectations

Workers increasingly expect AI skill development:

• Retention implications if you don’t invest
• Engagement benefits if you do
• Recruitment advantage

People costs are significant; investments that reduce turnover have value.

Start With Pilot ROI

If full program investment is hard to justify:

• Request pilot funding with clear success metrics
• Demonstrate returns on smaller investment
• Use results to justify scale-up

Pilots reduce risk and build the case progressively.

Common Budgeting Mistakes

Avoid these pitfalls:

Ignoring participant time. The largest cost is often people’s time. Acknowledge and plan for it.

Underestimating facilitation. Good facilitation makes learning effective but costs money. Budget adequately.

Forgetting ongoing costs. AI evolves; programs need refreshing. Budget for years two and beyond.

Over-centralising. Trying to do everything centrally is expensive. Leverage distributed resources.

Under-resourcing support. Training without follow-up support doesn’t produce results. Budget for sustained support.

Cutting contingency. Programs routinely cost more than planned. Protect contingency from cuts.

The Investment Mindset

AI upskilling is an investment, not an expense. Approached correctly:

• Returns exceed costs, often substantially
• Capability compounds over time
• Organisational adaptation enables future success

Budget with an investment mindset. Justify with investment logic. Measure investment returns.

The organisations that invest adequately in AI capability development position themselves for the future. Those that underinvest fall behind.

Budget accordingly.