Signal file: What happens when AI becomes your colleague?
From mass layoffs and vanishing entry-level roles to always-on-the-clock AI agents, this week's signals show how artificial intelligence is reshaping work faster than organisations are ready for
AI is disrupting traditional decision-making and ways of working at a pace that few organisations can keep up with. This has an knock-on impact on how organisation start to value labour – we’ve started to see large headcount reductions, restructured hiring pipelines, and new software that works through the night.
This week’s signals reveal a workplace where the rules around who does what, and what human work is needed for, are being rewritten from the ground up. The question organisations now face is not whether AI will change how they operate, but whether they are building the right conditions for people to thrive within that change.
Profitable companies cut workers to fund the AI
New data tracked by workforce analytics firm TrueUp shows US tech companies have cut more than 142,000 jobs in the first five months of 2026, a 33% increase on the same period last year, even as many of those same companies report record revenues. The defining feature of the current wave is that companies, including Meta, Amazon and Oracle, are not cutting because they are struggling. They are cutting to fund AI infrastructure, with four of the largest tech firms collectively committing around $700 billion USD in capital spending this year.
In action: As junior roles thin out across tech, organisations will need new models for building skills and growing the next generation of workers.
Entry-level accounting jobs are quietly disappearing
New research from BambooHR finds a third of new accounting and finance hires quit within their first year, with organisations now hiring senior staff at a ratio of three to one over entry-level roles. The underlying reason for this is that AI tools are allowing senior employees to absorb work that previously required junior support. What remains unclear is what entry-level roles are needed for now. BambooHR CFO Justin Judd explained that the new expectation is not data entry but the ability to think differently and orchestrate systems. The role has changed faster than the onboarding process has caught up.
In action: Organisations should be explicit about what they expect from junior hires in an AI-assisted environment, and build the development structures to match.
An AI agent that never clocks off joins your team
Microsoft introduced Scout at its Build 2026 conference, an always-on AI agent designed to work continuously inside Teams, Outlook and OneDrive without waiting to be asked. Unlike earlier assistant tools that respond to prompts, Scout monitors workflows, reschedules meetings, drafts messages and coordinates tasks in the background. Microsoft describes it as the first of a new category it calls Autopilots, AI systems that operate with their own identity and act on behalf of users within set permissions.
In action: Those organisations that are adopting autonomous AI tools will need clear governance around what these systems can see, decide and act on, and where human judgement still applies.
Attention is becoming the scarcest resource at work
New data compiled from multiple workforce studies, including Microsoft’s 2025 Work Trend Index and ActivTrak’s 2026 State of the Workplace report, shows the average focused working session has dropped to just 13 minutes. Employees now receive roughly 275 interruptions per day, one every two minutes, and research from UC Irvine finds it takes more than 23 minutes to fully regain focus after a single disruption. Despite AI adoption reaching 80% of employees, focus efficiency has fallen to a three-year low of 60%. The productivity gains from AI are real, but they are being eaten by the same fragmented, notification-heavy environments that AI tools were supposed to help solve.
In action: As AI handles more routine tasks, the ability to think deeply becomes more valuable. Organisations need deliberate strategies around protected time, meeting load and notification culture.
