AI, Applied
What changed. What matters. What to do about it.
Issue 02 · Fortnightly · 5-minute read
|
|
|
AI now has more access inside businesses than most leaders can account for.
Most businesses still treat AI as a yes/no decision: have we adopted it, or haven't we? That question matters less than it used to. Whether or not a business has formally rolled anything out, individual staff may already be using AI — often through their own personal accounts rather than anything the business set up. At the same time, AI features already built into everyday software are quietly gaining the ability to read, use and act on business data, whether or not anyone signed off on it.
|
|
The old question
“Have we adopted AI?”
|
→ |
The real question
“What is AI already doing inside our business — and who's watching it?”
|
|
|
|
So there's a widening gap between two different things: what a business has formally decided about AI, and what AI is already doing inside it. That gap is where the real risk — and the real opportunity — both live. Not in whether you've “adopted AI,” but in whether you actually know what's already running, what it can see, and who's responsible for it.
None of this means slowing AI down. It means having enough visibility and ownership to adopt it deliberately, rather than discovering after the fact how far it had already reached. The evidence below shows how fast that gap is opening — and what it takes to close it.
|
|
|
|
Three actions, not a checklist
|
1
|
Find out what's already touching your data.
Ask directly which AI tools, browser extensions or add-ons staff are currently using with company or customer information. Most leaders genuinely don't know — and that's the actual exposure.
|
|
2
|
Match oversight to what each tool can do — not one blanket policy.
A tool that only reads and summarises documents doesn't need the same sign-off as one that can send emails, update records or act on someone's behalf.
|
|
3
|
Assign clear ownership — not necessarily a title.
You don't need a Chief AI Officer. You need one person who can answer what AI tools are in use, what they can access, and who signed off on them.
|
|
|
|
|
|
Employee AI use tripled in a year — most of it outside any account you control
Verizon's 2026 Data Breach Investigations Report found regular AI use on corporate devices jumped from 15% of employees to 45% in twelve months, and 67% of those users are logging in through personal, non-corporate accounts rather than anything the business controls. Shadow AI — AI use nobody approved — is now the third most common non-malicious insider action in Verizon's data, roughly four times more common than a year earlier, and the single most common thing being fed into these tools is source code. The implication for a leader isn't really about breaches: it's that AI adoption inside your business may already be well underway, on accounts and devices you have no visibility into, regardless of any formal decision you've made.
Source: Verizon 2026 DBIR →
|
|
Gartner expects businesses to pull back a wave of AI agents next year — after they're already running
Gartner predicts 40% of enterprises will demote or decommission autonomous AI agents by 2027 because of governance gaps identified after production incidents. Its argument is that treating every agent the same creates problems: an agent that only reads information shouldn't require the same controls as one that can take actions or access sensitive systems. The lesson isn't “don't deploy agents.” It's that access and accountability need to be defined before deployment, not worked out afterwards.
Source: Gartner →
|
|
The Australian Government has started assigning clear ownership for AI
By the government's own 30 June 2026 deadline, 104 of the 106 Commonwealth agencies required to do so had appointed a Chief AI Officer — a specific person accountable for how that agency uses AI. Most businesses don't need a formal title like that. But the same underlying question applies: whether someone in the business can actually answer it when asked.
Source: ACS Information Age (Dept. of Finance) →
|
|
|
One Question For Your Business
Could you name who's accountable for every AI tool already used in your business?
|
|
|
|
|
|
|
That's it for this edition — short on purpose. If one of these changes what you do this week, it did its job.
|
|
Mia Pursehouse · Business Transformation Consultant
|
|
|
AI, Applied is an executive briefing for leaders who don't have time to follow AI but need to know what it means for their business.
|
|
AI transparency: AI tools assisted with research and drafting. Final selection, analysis and editorial review by Mia Pursehouse.
|
|
Read past issues
·
Manage preferences
·
Unsubscribe
Mia Pursehouse, Buttaba, NSW 2283
Sent to Reader
|
