This chapter follows four people through one year with the AI boss, from December 2026 to November 2027: Ingrid Halvorsen, Nordvik Region's HR director; Mikael Strand, who manages the Strandvik home-care unit; Jana Petersen, a claims team leader in Hansa Mutual's Hamburg office; and Thomas Brandt, who chairs Hansa's Konzernbetriebsrat. Parts IV to VI took one instrument at a time; here they run together in the order the year needed them. The year touches all five management functions: M1 Direction through rosters, routes and the allocation of claims; M2 Evaluation through dashboards and scores; M3 Consequence through hiring and pay inputs; M4 Development through the craft that deputies must keep; and M5 Care through travel time, rest and workload. The narrative is illustrative: the external law is observable, and the outcomes are expected consequences of the decisions recorded. Where earlier chapters gave conflicting dates, this chapter states the sequence the book follows.
In November 2026 neither organisation had a word for what it was governing. Nordvik Region had a procured rostering system, ShiftWise, about to draft home-care routes, a time-per-visit dashboard, assistants that put names against action items, and a recruitment-screening pilot that Ingrid Halvorsen had paused on 18 September (AIDR-0601) because the supplier could not show how its ranking was made (Chapter 7). On 16 November the regional director endorsed her commission to govern algorithmic management across the Region's about 6,000 staff, with a report to the regional executive committee due by 30 June 2027. The commission carries the first number of the range, AIDR-0600, because every other record is filed under it. Mikael Strand, with about 35 assistant nurses and care assistants, a third of them part-time or hourly, planned routes by hand; his unit was to be the first in home care to use ShiftWise routes.
At Hansa Mutual the AI boss was older and less visible. The claims-triage agent, a Mode 5 system, had allocated claims to individual handlers for years, and its throughput figures per handler reached team leaders under the group works agreement KBV-IT-2024. Jana Petersen led about 20 handlers in Hamburg and could reassign any claim the agent allocated; from January 2027 she would also direct two pilot agents under HM-AB-PILOT-01. Thomas Brandt had negotiated KBV-IT-2024 and sat on the European Works Council. The Dutch contestation CNT-HM-0001, decided on 2 November, had shown him a sales score that was advisory on paper and decisive in one unit (Chapters 8 and 27). On 23 November the executive board asked the chief HR officer and the chief risk officer for a group policy on AI in management, with a first report due on 30 April 2027 (AIDR-0650).
Primary topics explored
- Four people, two organisations
- Quarter one: commissions and agreements
- Quarter two: corrections, stops and the first reports
- Quarter three: oversight under pressure
- Quarter four: review and renegotiation
- What went wrong
About AI Boss
AI Boss is Book 7 of The Transformed Architecture. It examines what happens when artificial intelligence takes the manager's chair: algorithmic management of direction, evaluation, consequence, development and care, the effects on people, the management of AI agents, Europe's legal answer, and how to design an accountable AI boss, for European enterprises and public authorities.
It builds on the conclusions of Architecture as Code, AI Governance, AI Platform and Security, AI Red Team, AI Threat and The Transformed Security Architect.
Executives, line managers, HR and people leaders, works councils and trade unions, public employers, architects and AI governance leads, data protection and legal officers, procurement and programme leaders, and risk, compliance and audit.
Gunnar Nordqvist
Conceived, directed and published the books of The Transformed Architecture.
Continue with the full chapter
Chapter 30: A Year With the AI Boss: Nordvik Region and Hansa Mutual in AI Boss gives the detailed explanations, practical examples and implementation patterns behind this summary.
This whitepaper belongs to “Experience and Verdict” and is one of 37 drawn from the manuscript.