| Last Week in HR AI: Issue #15 | Week of October 5, 2026 |
1,950 hours a year off a six-person HR advisory team
The hardest part of the build was writing down what the team knew but had never documented.
|
1 TechTarget2 healthsystemcio.com3 bbrw.com4 mk.co.kr
Ikona's Take on the past week (October 5, 2026) in HR + AI
|
IO
|
Ian O'Keefe
Co-founder & CEO, Ikona Analytics
|
The easy assumption is that hours freed by AI are savings waiting to be booked. The week's evidence says that is the wrong ledger. AI is freeing real capacity inside functions, and the hard decision is where those hours go: firms that cut first are rehiring, firms that kept their people cannot yet show the growth, and the ones getting it right redeploy staff on top of clean data, a usable skills map, and an integrated stack. Robert Half found that just under a third of hiring managers who cut roles mainly because of AI later rehired the same or similar positions. Oversight ran heavier than planned, and institutional knowledge left with the people.
Intermountain Health shows the alternative. Once agents matched or beat 40 informaticists in selected areas, most of those staff moved to at-the-elbow support, and the readiness signals were unglamorous: clean history and clear ticket ownership. OneDigital says AI now handles roughly 30% of the work in its largest job families, yet it cannot measure whether that capacity reaches retention or new sales. That is the first gap your CFO will find. Redeployment needs somewhere to land, and as Deloitte warns, without orchestrated data, agents just process the existing mess faster.
Before the next pilot, ask one question: can you name the work those hours will go to, and the measure that will prove it paid off? Reply and tell me where yours are going. If the honest answer is "nowhere yet," we'd welcome the conversation.
| The six to read | The stories that matter for the Office of HR — with our take on each. |
Robert Half research shows that just under a third of US and Canadian hiring managers who eliminated roles mainly because of AI later brought back the same or similar positions. The leading reasons were oversight and quality-control demands that ran heavier than expected, along with the loss of relationship skills and institutional knowledge. Nicole Gorton of Robert Half expects Australia to follow gradually and recommends planning the workforce at the task level. Commonwealth Bank and IBM are named as earlier examples of companies reversing course.
Also covered by: ibtimes.co.uk, digitaljournal.com
Over three years, OneDigital has built AI coworkers tailored to more than 20 of its largest job families, and the company estimates they now take on roughly 30% of the work in those roles. Leaders report that adoption picked up after they positioned the AI as a teammate instead of a tool. Executives acknowledge they cannot yet measure whether the freed capacity is turning into client retention or new sales.
During an Epic go-live that folded nine electronic health record platforms into one, Intermountain Health used AI agents to triage support tickets, according to its chief digital officer. After a week of passive testing, the agents performed as well as or better than 40 informaticists in selected application areas, and most of those staff shifted to at-the-elbow support. Six months later, the agents were assisting with about 63% of service desk incidents. He identifies four readiness signals, among them clean history and clear ticket ownership.
|
From Ikona
Freed hours need somewhere to land
We diagnose how work and knowledge actually flow through your HR function and capture the judgment that lives in people's heads. Then we turn it into a structured asset your AI and your redeployment plans can stand on. Talk to us about yours → |
Chemist Warehouse is turning AIHRA, the HR-inbox agent it built with Microsoft partner Insurgence, into a reusable pattern for other teams following a 10-week pilot in HR. The agent drafts responses to routine staff questions and is estimated to save 1,950 hours a year for an advisory team of roughly six people. The company says the hardest part of the build was writing down tacit knowledge that had never been documented.
Shinhan Financial Group is piloting AI persona agents designed to absorb the judgment and experience of department heads so their know-how stays put when managers rotate. In 11 units spanning Shinhan Bank, Card, Investment, and Life, each department head works with a junior, AI-capable staffer to build the agent. The agents are meant to flag concerns a manager would likely raise on a report and propose responses based on past decisions. Chairman Jin Ok-dong launched the program himself; prototypes are under review through November, with a possible group-wide rollout in 2027.
Samsung Electronics' Device eXperience Division has rolled out My Skills, a career platform that organizes employees by skill instead of by job title. AI extracts skills from each person's work history, employees can edit the results, and the system suggests training for gaps and matches people to internal roles and projects. The framework behind it was built from more than 1.12 million data points and contains 10,487 skills across 16 job categories, tied to 2,187 courses. Samsung has not reported outcome results.
Argues that the CHRO role has moved from administering policy to acting as a strategic business partner, and that a significant gap persists even as organizations invest heavily in HR technology.
The smallest detail this week may be the most useful: Chemist Warehouse said the hardest part of building its HR agent was documenting knowledge nobody had ever written down. Set that next to the Robert Half finding, where roles came back because institutional knowledge left with the people, and Shinhan pairing department heads with junior staff to capture their judgment before they rotate out. The order of operations is clear: capture what your people know before you decide what their freed hours are worth. On Monday, pick the one role your next pilot touches and find out whose judgment never made it into the ticket history.
If your team is wrestling with workforce intelligence, AI readiness, or a transformation that has to land, we'd love to help. Reach out and tell us what you're working on.
Get in touch →Last Week in HR AI
Practitioner perspectives and diagnostic frameworks, delivered the week they publish. Be the first to know when new research drops.
No spam, ever. Unsubscribe anytime.