AI systems that do more than give answers: they take steps in a work process on their own, such as assessing an application, answering an email or passing on a ticket. That makes them actors rather than tools, and they need an owner, a mandate and an off switch.
Algorithmic management is spreading from platform work to warehouses, transport and shops. A Swedish study of 978 drivers and warehouse workers links heavy algorithmic management to twice the rate of mental health complaints. Brussels and The Hague are drafting rules.
Publications per month for Algorithmic management in warehouses, delivery and retail
The number of AI agents is growing fast, but Gartner expects over four in ten agent projects to be cancelled by the end of 2027 because of rising costs and unclear value. Vendors, meanwhile, are sticking the agent label on existing products. The value goes to companies that manage agents as a single programme.
Publications per month for Agent-washing and the cost and value of agent projects
Whether employees use AI and experience it as an improvement depends mainly on whether their own manager visibly supports it. At the same time, half of HR directors doubt their managers can do this, and training alone does little as long as those managers are overloaded.
Publications per month for The manager's role in how employees use AI
Managers use AI almost twice as often as their staff and think the direction is clear. The workforce does not experience it that way. The difference lies in what a manager visibly does: leading by example, setting a quality standard and saying what happens to the time saved.
Publications per month for AI use and AI strategy as seen by leaders and by staff
More than half of nurses, carers and support workers use AI, mostly on their own initiative and without training. Hospitals are adopting AI policies at speed, but half of healthcare professionals see no clear policy and two thirds feel insufficiently prepared.
Publications per month for AI use, training and AI policy among healthcare professionals
The share of developers willing to leave application logic to AI rose from 18 to 50 per cent in a year. Measuring the return no longer works: METR cannot find enough developers still willing to work without AI, and 66 per cent feel metrics miss their contribution.
Publications per month for AI in software development and the trouble with measuring productivity
We use Google Analytics to see which pages people read, so we can improve the site. This requires cookies. You can change your choice at any time via Cookies at the bottom of the page. More in our privacy statement.
