More media organizations are creating dedicated AI functions. The question I hear constantly: what roles should we hire?

The answer depends on your size, ambitions, and existing capabilities. But after watching many organizations build AI teams—some successfully, some not—I can offer guidance on what tends to work.

The Essential Role: AI Lead

Every organization investing in AI needs someone to own it.

This person’s responsibilities include: evaluating tools and opportunities, developing policies and guidelines, coordinating implementation across teams, troubleshooting problems, and connecting AI strategy to business objectives.

The title varies—AI Editor, AI Strategy Lead, Head of AI Innovation—but the function is consistent: a single point of accountability for AI initiatives.

What to look for:

• Editorial background combined with technical literacy
• Strategic thinking—ability to connect AI capabilities to business value
• Change management skills—getting journalists to adopt new tools
• Communication ability—translating between technical and editorial stakeholders

Where to find them:

The best candidates often come from within. Someone who’s already experimented with AI tools, understands your organization’s culture, and has credibility with editorial staff may outperform an external hire with impressive credentials.

If hiring externally, look at journalism-school programs with AI focus, tech companies with editorial partnerships, or consultants who’ve worked with multiple media organizations.

Salary expectations:

Varies dramatically by market and organization size. In Australia, expect ranges from $100,000 for mid-level roles at smaller publishers to $180,000+ for senior positions at major organizations.

The Technical Support: AI Engineer/Developer

If you’re building anything custom—not just implementing off-the-shelf tools—you need technical expertise.

An AI engineer or developer can: build integrations between AI tools and your systems, develop custom applications for your specific needs, evaluate technical aspects of tool selection, and solve problems that vendor support can’t address.

What to look for:

• Software development skills, particularly Python
• Understanding of AI/ML concepts (doesn’t need to be a specialist)
• Experience building integrations and working with APIs
• Interest in media and journalism applications

Where to find them:

This is a general engineering role with media application, not a specialized AI position. Traditional tech recruitment channels work. Emphasize the interesting problem space—many engineers find media applications more compelling than generic enterprise work.

Salary expectations:

Engineering salaries at media companies typically lag pure tech, but not dramatically. Expect $120,000-180,000 for competent engineers in most Australian markets.

The Training Function: AI Educator

For larger organizations, someone focused on training and education often proves valuable.

This person develops and delivers training programs, creates documentation and resources, and provides hands-on support for staff learning new tools.

What to look for:

• Strong communication and teaching skills
• Practical experience with the AI tools you’re using
• Patience with varying technical comfort levels
• Ability to create accessible training materials

Where to find them:

Often internal promotion works best—someone who’s learned the tools and enjoys teaching others. Traditional L&D backgrounds combined with AI literacy can also work.

Salary expectations:

Training roles typically pay less than strategic roles—$70,000-100,000 in most markets.

The Cross-Functional Champions

Beyond dedicated roles, identify AI champions across functions.

These aren’t full-time AI positions—they’re existing staff who become go-to resources for AI questions within their teams. A reporter who’s skilled with AI research tools, an editor who understands AI content policies, a producer who knows AI transcription workflows.

Champions don’t need new titles or significant salary increases. They need recognition, time to stay current, and connection to central AI leadership.

This distributed model scales better than centralized expertise and builds AI capability throughout the organization rather than siloing it.

Consulting and External Support

Not everything needs to be hired permanently.

For specialized projects—building a custom tool, conducting a major evaluation, developing strategy—external consultants may make more sense than permanent hires.

When to use consultants:

• One-time projects with clear scope
• Specialized expertise you don’t need ongoing
• Strategic advice requiring outside perspective
• Capacity overflow during implementation phases

Media-focused AI consultancies have emerged to serve this need. General AI consultancies can work too, though they may lack industry-specific context. For implementations that touch newsroom operations, firms with journalism background tend to produce better results. Firms specialising in custom AI development understand media-specific requirements that general technologists often miss.

Team Structure

How should AI roles relate to existing structure?

Option 1: Centralized AI team. A standalone group reporting to senior leadership, serving all parts of the organization. Pros: clear accountability, consistent approach. Cons: can feel disconnected from operational realities.

Option 2: Embedded in editorial. AI roles report into editorial leadership, sitting alongside newsroom functions. Pros: close to where work happens, editorial credibility. Cons: may miss non-editorial opportunities.

Option 3: Hybrid. Strategic leadership centralized; execution distributed. An AI Lead sets direction while champions and specialists work within teams. Pros: combines benefits. Cons: coordination complexity.

For most media organizations, the hybrid model works best—provided the central AI Lead has authority and relationships across the organization.

Sequencing Hires

If you’re starting from scratch with limited budget, how should you sequence?

First hire: AI Lead who can develop strategy, evaluate tools, and begin training staff. This person will determine what else you need.

Second hire (if building custom): Technical support to build integrations and custom applications.

Third hire (at scale): Training/education focus as implementation expands.

Ongoing: Develop champions across teams as part of broader AI literacy efforts.

Many organizations can start with just the first hire—a capable AI Lead who leverages existing resources, vendors, and consultants for other needs.

What Not to Do

Common mistakes in AI team building:

Over-hiring early. Don’t build a large team before knowing what you need. Start small and scale based on demonstrated value.

Pure technologists without editorial understanding. AI in media requires understanding journalism, not just algorithms. Technical skills without editorial grounding produce disconnected solutions.

Siloing AI. AI should enhance journalism, not exist separately from it. Keep AI roles connected to editorial operations.

Hiring hype over substance. AI is fashionable; some candidates talk better than they execute. Test practical capabilities, not just conceptual fluency.

The Talent Opportunity

Here’s the good news: media AI roles are genuinely interesting.

The problems are compelling—how do we maintain journalism values while gaining AI efficiency? How do we use technology to do better journalism, not just cheaper journalism?

Talented people are drawn to interesting problems. Media AI roles can attract strong candidates who want meaningful work, not just good compensation.

Emphasize the mission. The right candidates will care about journalism’s future, not just AI’s potential.

Build your team thoughtfully, and you’ll attract people worth having.