50%
of respondents use AI daily for work
report · edition 2026 · australia
What 2,400 workers told us about how they actually use AI on the job — quietly, daily, and mostly without permission. A field report on the largest workforce shift in a generation.
2,418
respondents
14
industries
6
key themes
jul '26
fielded
01 — At a glance
87%
say AI has made them more efficient, meaningfully or in specific tasks
7%
say their organisation is measuring AI impact with clear metrics or KPIs
75%
see AI as both a productivity opportunity and a policy risk
02 — Who we heard from
A representative cut across location, seniority and industry — offering meaningful insight into how AI is showing up and being managed in teams and organisations today.
Location
943
respondents
- Australia62%
- New Zealand28%
- United States10%
Seniority
44%
People leaders
- C-suite11%
- Director14%
- Head of19%
- Manager26%
- Individual Contributor30%
Industry
Technology
22%
Professional Services
17%
Financial Services
13%
Government
9%
Health
8%
Education
7%
Retail & eCommerce
6%
Media & Entertainment
5%
Telecommunications
5%
Energy & Resources
4%
Logistics & Transport
4%
Note: This survey closed on 12 July 2026 and captures a moment in time in a rapidly developing space.
03 — Adoption & Maturity
Most workers are using AI regularly and many organisations have moved beyond ‘not started’ — but maturity is uneven, with the majority of organisations sitting at partial implementation.
AI literacy is now showing up in roles, hiring conversations and candidate expectations — fluency with these tools is becoming a baseline workplace skill.
TBD
Talent Leader
Workplace adoption almost always starts with individual productivity. The harder, slower work is turning that into true transformation across systems and teams.
Jack Jorgensen
General Manager — Data, AI & Innovation
50%
Use AI daily for work
29%
Sit at the most common org-level stage: implemented in some functions or workflows
10%
Say AI is embedded in business strategy and operating model
Organisation AI maturity
2025 vs 2026
Restricted / Not started
8%
Informal experimentation
22%
Pilots
24%
Implemented in workflows
29%
Embedded in strategy
17%
Personal AI maturity
At work vs outside work
Rarely or never
9%
Quick answers / research
24%
Stronger outputs
27%
Repeatable workflows
20%
Connected to tools / systems
12%
Across roles / functions
8%
04 — Productivity & ROI
Productivity is strong at the individual level — most respondents say AI has made them more efficient. But that hasn't translated cleanly into measurable organisational ROI. Organisations are using AI faster than they can measure its impact.
81%
Say their organisation's AI rules are clear
38%
Are not confident people understand what data can and cannot be entered into AI tools
20%
Say their organisation has never provided AI training or a mandatory policy refresh
Has AI made you more efficient at work?
Saved meaningful time
41%
Small / specific task gains
46%
No noticeable difference
8%
Created errors / rework
3%
More expectation, same workload
2%
Is your organisation formally measuring AI impact?
7%
Clear KPIs
- Yes, clear metrics or KPIs7%
- Yes, but informal22%
- Starting to define28%
- No, but plan to21%
- No current plans14%
- Not sure8%
Metrics used to measure AI impact
Time saved
64%
Productivity / output
42%
Quality of work
36%
Cost reduction
30%
Adoption / usage rates
27%
Speed of delivery
24%
Employee experience
19%
Error reduction
17%
87%
Report some form of efficiency gain
7%
Say their organisation has clear metrics or KPIs
16%
Say their organisation has clearly seen measurable ROI
Time saved is a useful starting point, but real AI measurement should also consider quality, risk, adoption, workflow redesign and business outcomes. Most orgs are stuck measuring inputs, not impact.
Jack Jorgensen
General Manager — Data, AI & Innovation
05 — Policy & Risk
AI is being used as a productivity tool, but governance isn't keeping pace. Many say the rules are clear — yet confidence around data handling is mixed. The risk isn't deliberate misuse, it's everyday uncertainty about what can be entered, what needs checking, and where judgement is required.
Policy clarity vs data confidence
Q15 — rule clarity · Q17 — data confidence
Rules: very clear
38%
Rules: somewhat clear
43%
Confidence: very
22%
Confidence: somewhat
40%
Confidence: not very
28%
Confidence: not at all
10%
Last AI training or policy refresh
Past month
9%
Past 3 months
17%
Past 6 months
21%
Past 12 months
18%
More than 12 months
11%
Never
20%
Not sure
4%
Highest-risk AI behaviours
Top 5
Entering confidential data
58%
Relying on outputs unchecked
49%
Decisions without human oversight
41%
Using unapproved AI tools
36%
Ignoring internal AI policies
27%
Policy on paper isn't the same as policy in practice. Real adoption needs training, manager enablement and clear escalation paths — not just a published document.
Sarah Blanchard
Head of Talent Advisory
Data risk isn't theoretical anymore. Policy has to be operationalised inside workflows and tools — otherwise the gap between intent and behaviour keeps widening.
Jack Jorgensen
General Manager — Data, AI & Innovation
06 — Trust & Quality
In the age of ‘AI slop’, 49% believe AI output quality is improving — but trust is conditional. Top concerns are loss of human judgement, data security, over-reliance and accuracy. The conversation has matured from ‘can AI do this?’ to ‘when should we trust it, and where do humans stay in control?’
31%
Say output quality depends heavily on the tool or use case
21%
Are concerned about the loss of human judgement
50%
Say their organisation has become somewhat more realistic — but hype still drives decisions
Is AI output improving or declining?
49%
Improving
- Improving49%
- Depends on tool / use case31%
- About the same12%
- Declining8%
Biggest concerns about using AI at work
Top 6
Loss of human judgement
21%
Data security
19%
Over-reliance
17%
Accuracy / hallucination
16%
Quality of output
14%
Ethical / bias
11%
Are organisations becoming more realistic?
Much more realistic
14%
Somewhat more realistic
50%
About the same
22%
Still hype-driven
11%
Not sure
3%
AI's limits are now visible to everyone using it. Hallucination, brittle outputs and use-case mismatch are why QA and judgement matter more, not less, as adoption grows.
Jack Jorgensen
General Manager — Data, AI & Innovation
07 — Managers
Managers sit at the intersection of adoption, quality control, productivity expectations, policy enforcement and team behaviour. Most feel at least somewhat prepared, but the challenges they identify show how much complexity now sits with the management layer.
77%
Of managers feel prepared to manage AI use within their team
37%
Say the hardest part is ensuring data security and compliance
42%
Say AI will change the role of managers, but not reduce demand
Manager preparedness
Getting better
38%
Depends on tool / use case
28%
About the same
19%
Getting worse
7%
Don't use AI enough to say
8%
Hardest parts of managing AI use
Top 5
Data security & compliance
37%
Knowing when judgement is required
31%
Managing overuse / poor use
24%
Measuring productivity gains
22%
Setting expectations for quality
19%
Expected impact on middle management
42%
Role changes
- Role changes, demand stays42%
- More important18%
- Pressure without support17%
- Some roles reduced13%
- Little impact6%
- Not sure4%
Middle management is where transformation either takes hold or stalls. They're the layer turning policy and tools into actual team behaviour — and they need enablement, not just expectation.
Cameron Robinson
Head of Enterprise
08 — Workforce Planning
AI is beginning to shape workforce planning, but the dominant response is upskilling rather than redesign. Organisations aren't universally changing hiring strategies or operating models — many are still watching, planning or experimenting.
49%
Say their organisation is upskilling existing employees
24%
Say their organisation is redesigning roles or workflows
21%
Say their organisation is hiring for AI specialist roles
How AI is changing workforce planning
Upskilling existing employees
49%
Redesigning roles / workflows
24%
Hiring AI specialists
21%
Reducing manual / repetitive work
19%
Changing skills hired for
17%
Using AI in hiring
12%
External partners / consultants
10%
Not yet changing
18%
From upskilling to restructuring
Grouped response themes
Capability building
58%
Work redesign
34%
Hiring strategy
24%
External support
10%
No change yet
18%
The hiring signal is shifting. The premium isn't on people who use AI — it's on people who can apply it critically, validate outputs, and redesign how work gets done.
TBD
Talent Leader
Skills alone don't create transformation. Workflows, governance and systems have to be redesigned around AI capability — otherwise upskilling just produces faster individual users, not better operating models.
Jack Jorgensen
General Manager — Data, AI & Innovation
09 — So what?
Five reads on the data — by the people who'll need to act on it.
For Business leaders
AI adoption is not the milestone. The next challenge is proving impact. Leaders need to define what success looks like, decide what should be measured, and make sure productivity gains do not quietly become unmanaged workload increases.
For Hiring managers
AI is changing what capability looks like. The value isn't simply whether someone can use AI, but whether they can use it critically, safely and effectively. Hiring processes may need to test judgement, workflow thinking, data awareness and the ability to validate outputs.
For HR, P&C and TA leaders
The grey zone is now a people-strategy issue. Policies, training and governance need to be practical enough for employees and managers to apply in real work. AI adoption will not scale on policy alone.
For Technology & transformation leaders
The data points to an execution gap. People are using AI, but many organisations are still defining measurement, governance and workflow integration. The opportunity is to move from tool access to operating-model impact.
For Jobseekers and contractors
AI literacy is becoming a workplace expectation, but the strongest signal isn't just using AI — it's knowing when to use it, how to check it, and how to apply it to improve quality, speed and judgement.
Get the full dataset
Industry breakdowns, regional splits, and crosstabs by role and tenure are available on request. Talk to the team about what the data says for your sector.