Happy Wednesday, everyone. I'm Frank Richardson, an organisational psychologist observing the workplace with curiosity and care. Each week, I share insights to help HR leaders better understand the people behind the processes and build cultures where both individuals and organisations can thrive.
This week in workplace whiplash 🌀
AI had another big week, but not for the reasons many organisations would have hoped:
🤖 Microsoft Cuts Another 4,800 Jobs
Microsoft is laying off around 4,800 employees, including approximately 3,200 roles across Xbox, as part of a major organisational restructure. While the company says the cuts aren't driven by AI alone, it acknowledged that AI is changing how work gets done and that employees will need to continue building new skills as roles evolve.
👉HCA Mag⚖️ The Workday Lawsuit Just Got More Serious
A US federal judge has allowed key parts of the landmark lawsuit against Workday's AI-powered hiring tools to proceed. The case centres on claims that the platform's automated screening discriminated against applicants based on age, race and disability. Whatever the eventual outcome, one thing is becoming clear: "the algorithm did it" is unlikely to be a successful legal defence.
👉SHRM👩⚖️ HR Leaders Could Be Personally Liable for AI Decisions
Employment lawyers are warning that HR professionals themselves could face personal legal liability if AI-driven hiring, leave management or workforce decisions result in discrimination or breaches of employment law. The message is simple: AI can support HR decisions, but it can't replace human accountability.
👉HCA Mag
Speaking of AI... this week's article explores another interesting consequence of bringing it into our workplaces.
🤝 SPONSORED BY METAVIEW
Two years ago, AI agents started shipping production code. Engineering teams stopped scaling through headcount and started scaling through leverage. Now the same shift is reaching recruiting.
Metaview's new autonomous agent, Fillmore, researches candidates, writes personalised outreach, manages follow-ups, and books screening calls on its own. You tell it who you're looking for. It handles the rest.
Key Takeaways:
🤖 Outbound recruiting that runs itself: Fillmore sources candidates, writes personalised outreach, manages follow-ups, and schedules calls end-to-end with no human coordination required
🚀 You delegate outcomes, not tasks: Give Fillmore a role and it delivers conversations, operating across email, LinkedIn, and WhatsApp while your team focuses on hiring
💬 No templates, every message is personalised: Outreach is written specifically for each candidate, not pulled from a library
🎯 Already making hires: Cognition, Rippling, Luma, and Replit are testing it, and Luma AI has reportedly already made hires sourced through Fillmore
Join the waitlist 👉 metaview.ai/fillmore
Yesterday, I was completely stuck on a piece of writing. I opened ChatGPT and ten minutes later I'd untangled the structure, rewritten a couple of paragraphs and moved on with my day.
Five years ago, I'd have done something completely different. I'd have wandered over to a colleague's desk and said, "Can I run something past you?" That conversation almost certainly wouldn't have stayed on the original topic. We'd have ended up talking about another project, sharing an idea one of us hadn't considered, or disappearing down a completely unrelated rabbit hole.
ChatGPT gave me exactly what I needed. But it also took away a conversation I'd almost certainly have had with another person.
And turns out I’m not the only one. Recent research suggests 37% of employees have already used AI for companionship at work, while almost three-quarters have used it for advice or brainstorming. At the same time, employers are beginning to report growing concerns about isolation as AI becomes embedded into everyday work.
One conversation hardly seems worth noticing. I’m sure most of us would happily trade a quick trip to a colleague's desk for ten extra minutes in the day. But workplaces aren't built on one conversation…. they're built on thousands of them.
These days, what's your first instinct when you're stuck on a work problem?
🧠The behavioural science lens
As AI becomes our first instinct for more and more tasks, it changes some surprisingly important workplace behaviours:
Convenience changes relationships: Behavioural scientists have known for decades that humans gravitate towards the easiest available option. Asking AI is fast, private and effortless, so it's hardly surprising that many of us instinctively open ChatGPT before turning to another person. But over time, convenience changes who we rely on and how workplace relationships are formed.
The smallest conversations often have the biggest impact: Sociologist Mark Granovetter's research on weak ties showed that many of our most valuable professional relationships are built through brief, everyday interactions rather than our closest friendships. A quick "Have you got two minutes?" often becomes mentoring, collaboration or an idea neither person expected. Every time AI answers that question instead, one small opportunity for connection disappears.
Trust grows through repetition: Psychological safety isn't created by a single workshop or a leader announcing an open-door policy. It develops through hundreds of ordinary moments where people ask for help, admit uncertainty and test unfinished ideas with one another. If AI becomes our default sounding board, we're still getting answers, but we're spending fewer moments building trust with the people we actually work alongside.
Your best colleague rarely answers the question you asked: One of AI's greatest strengths is that it responds directly to your prompt. Colleagues rarely do. They challenge your assumptions, connect your problem to something completely different or share an experience that changes how you think about it. Research on transactive memory systems suggests those exchanges are one of the reasons high-performing teams consistently outperform groups of equally talented individuals working in isolation.
🚀What this means for leaders
As AI becomes embedded into everyday work, leaders have an opportunity to think more deliberately about which workplace interactions they want technology to accelerate and which ones they want people to keep having:
Protect conversations with hidden value: Many workplace conversations achieve far more than answering a question. They build trust, uncover expertise elsewhere in the business and strengthen relationships that people rely on later. Those benefits are easy to overlook because they emerge naturally rather than being the purpose of the conversation.
Watch where people go for help: The first place employees seek advice says a lot about an organisation's culture. Frequent use of AI may simply reflect a brilliant tool, but it's also worth paying attention to how often colleagues learn from one another. Patterns in help-seeking reveal how knowledge moves through an organisation.
Measure connection alongside productivity: Most AI success stories focus on faster work, lower costs and greater efficiency. Equally valuable indicators include mentoring, collaboration and informal knowledge sharing. The organisations that thrive with AI will pay attention to both sets of measures.
💬 Final thoughts
AI might become the easiest colleague you'll ever have. It answers immediately. It never tells you it's too busy. It never makes you feel dumb for asking the same question twice.
The trouble is, your best colleagues were never valuable simply because they had the answers. They were valuable because one question somehow turned into five ideas and a completely different way of looking at the problem.
Work has always run on those small, forgettable conversations. And I'd hate for us to realise how important they were only after they'd disappeared.
How's the depth of today's edition?
If something here speaks to you, I’d love to hear it.
Until next week,
Frank
P.S. If you want to get a feature about your own story, reply to this email. If you’d like to reach our newsletter audience (founders, creators, and marketers), click the button below.
If you’re new here, I’m over the moon you’ve joined us! To help me craft content that’s actually useful (and not just noise in your inbox), I’d love it if you took 1 minute to answer this quick survey below. Your insights help shape everything I write.
✨ Insane Media is more than one voice
💡 Dive into our other newsletters - where psychology meets the founders, creator economy, e-commerce marketing, and AI founders.
