Head of AI and Chief AI Officer recruitment, done with evidence.

The Galileo salary dataset benchmarks a Head of AI at large Sydney companies at a $291,000 median total package including super, with a band of $248,000 to $335,000 at principal level (source: Galileo salary dataset, July 2026, built from our market research, live tracking of advertised Australian AI roles, and validated placement data). Company size moves the number: APRA-regulated and enterprise employers pay above large-company rates, while mid-market firms pay below them. Advertised packages across the broader market run $230,000 to $350,000, with the top-end figures usually quoted as total package and the rest as base salary excluding super, which adds 12% (market observation, July 2026). Live mandates Galileo Search is briefed on in Sydney and Melbourne sit at $260,000 to $340,000 base plus super (source: Galileo Search live mandate data, 2026). Scope moves the number too: a Head of AI who owns a P&L outcome earns like an executive; one who leads a single squad earns like a principal engineer with a bigger title.

The difference is altitude and reporting line, not the technology. A Chief AI Officer sits on the executive committee and owns AI strategy, governance and value across the whole enterprise. A GM of AI runs AI as a business unit with its own budget and number. A Head of AI leads the function and its delivery, usually reporting to a CTO, CDO or COO. Match the title to the mandate you can actually give: hiring a CAIO you will not empower fails faster than hiring no one.

Plan for three to five months from brief to start date for a permanent hire. US benchmarks put a well-run Chief AI Officer search at 19 to 22 weeks from kickoff to day one (source: MSH, https://www.talentmsh.com/insights/hire-a-chief-ai-officer); Australian searches run similar because the talent pool is smaller and almost entirely passive (directional estimate, method: benchmark plus AU market read). If the business cannot wait, start a contract or fractional AI leader while the permanent search runs.

Yes, in most cases hire the leader first. 70% of the AI challenge is people and process rather than technology (source: BCG 10-20-70 rule, via Galileo evidence review 2026), so engineers hired before anyone owns strategy tend to build pilots that stall: 95% of GenAI pilots show no P&L impact (source: MIT NANDA, The GenAI Divide 2025, via Fortune, https://fortune.com/2025/08/18/mit-report-95-percent-generative-ai-pilots-at-companies-failing-cfo/). The exception is a company with one narrow, well-defined AI use case, which can start with a senior engineer or architect.

A strong first-100-days pattern: days 1 to 30, inventory the current state (systems, data, shadow AI use, spend); days 31 to 60, build a costed option set against business outcomes; days 61 to 90, put decisions to the executive and lock the roadmap and operating model (source: Galileo Search first-100-days framework, built from CAIO tenure research 2026). Test candidates against this in interview: ask exactly what they would do in the first 30 days and listen for inventory before ideas.

Most Australian mid-market companies (100 to 5,000 staff) need a Head of AI or GM of AI, not a C-suite CAIO. Dedicated AI executives now exist in roughly 26 to 38% of large enterprises with an average team of five (source: CAIO market research 2025-2026, via Galileo evidence review; primary citations available on request). The pressure is real either way: Gartner projects 75% of data and AI executives who are not seen as essential will lose the seat by 2027 (source: Gartner, via Galileo evidence review 2026), so define the mandate and the measure of success before you define the title.

Not on job boards. Leaders at this level are almost all employed and passive; they move through direct approach, referral and trusted search relationships (consistent with hire-guide consensus, for example https://www.kore1.com/how-to-hire-head-of-ai-2026/). The realistic AU pools: AI and data leaders inside consultancies wanting an owner's seat, VP-level product and platform leaders who have shipped ML at scale, returning expats, and internal Head of Data promotions. A specialist search maps all four pools before approaching anyone.

The Head of AI owns what gets built; the AI Transformation Lead owns whether the organisation actually adopts it: change strategy, enablement, adoption measurement and the operating rhythm. Roughly half of AI initiative failures trace to change management rather than technology, and structured change management is associated with several times better outcomes (source: Prosci research, via Galileo evidence review 2026). Larger or more federated organisations often need both roles; a mid-market firm usually starts with one leader wearing both hats.

As of mid-2026 LinkedIn lists around 195 Head of Artificial Intelligence roles in Australia (source: https://au.linkedin.com/jobs/head-of-artificial-intelligence-jobs, checked 2026-07-06; counts change daily). Galileo's own scrape of Sydney and Melbourne postings found Head of AI/GM AI mandates in the low twenties, versus 111 live AI Engineer postings (source: Galileo internal demand scrape, May 2026, refreshed monthly). Leadership demand is real but thin, which is exactly why most searches come down to approaching people who are not looking.

Contract (or fractional) suits three situations: you need momentum while a permanent search runs, you want the strategy and first roadmap built before committing to a permanent leadership package (a Head of AI benchmarks at a $291,000 median total package including super at large Sydney companies, principal level, Galileo salary dataset, July 2026), or the mandate is genuinely 6 to 12 months of setup. Permanent is right once the mandate is enduring: owning the roadmap, the team and a number. Many Australian companies sequence both: a contract AI leader stands up the function, then helps hire their permanent successor. Galileo Search runs both contract and permanent searches.