Eye on the Horizon is our monthly look at some of the emerging shifts and undercurrents shaping the future of work. Clients often ask us to make sense of where things are heading, not the trends everyone’s already talking about, but often it's the quieter signals that tend to matter more, precisely because they haven’t hit the mainstream yet. Strategic foresight isn’t about predicting the future, it’s about noticing change early enough to do something useful with it, before it becomes the story everyone else is reacting to. So here are four shifts that made me sit up and take notice this month, along with what I think they mean for your business, your team and your leadership.
1] The perils of ‘trendslop’ – when chasing the trend becomes the strategy
I recently came across a term that stuck – ‘trendslop’, which describes the flood of AI-generated commentary now dressed up as insight. It’s an emerging issue for anyone trying to make sense of where work is heading.
The term itself is only a few months old. It was coined in March 2026 by Angelo Romasanta of Esade Business School in Barcelona, working with colleagues at the University of Sydney and NYU Stern, in a piece of research for Harvard Business Review. They'd been testing what happens when you ask leading AI models for strategic advice, and found a consistent pattern: the models kept recommending whatever aligned with the trendiest management buzzwords, rather than whatever actually fitted the situation in front of them. They called that bias "trendslop."
It's a close cousin of "workslop", the term BetterUp Labs and Stanford's Social Media Lab coined in 2025 for AI-generated work output that looks polished but is hollow underneath, and which ends up creating more work than it saves. But the two point at different failure modes. Workslop is about the output: content that looks finished but isn't. Trendslop is about the input: a bias, in both the AI models and the humans prompting them, towards whatever's fashionable over whatever's actually right for the business in question. You can produce trendslop without any AI in the loop at all, simply by choosing the trend everyone's talking about over the one that fits your circumstances. Put them together and you get a fairly bleak picture of AI-assisted strategy: buzzword-driven advice in, unhelpful polished output out.
Analysis published in July 2026 by AI-detection firm Pangram, covering more than a million posts, found that over 40% of long-form LinkedIn posts are now fully machine-written, with LinkedIn alone accounting for nearly two-thirds of all AI-generated content across the platforms it studied.
That matters because a huge amount of the ‘trend’ content circulating in leadership and HR circles right now, the reports, the LinkedIn posts, the ‘10 trends for 2027’ listicles, was never actually thought through by a person. Gartner’s own 2026 Hype Cycle for Agentic AI shows where that leads in practice: organisations are greenlighting AI projects on the strength of a slick demo rather than a proper business case, and Gartner now predicts that more than 40% of agentic AI projects will be cancelled by the end of 2027 as a direct result.
Trendslop is the by-product of being over-reliant on trend-chasing and letting that quietly replace genuine strategic thinking. Leaders make real decisions, on talent, on where to focus investments, on restructuring, based on a content ecosystem where a growing share of the ‘evidence’ feeding them is recycled, unattributed and dressed up as analysis. Chase the buzzword and skip the scrutiny, and you inherit someone else’s hype cycle – and its inevitable comedown, along with it.
The fix isn’t complicated, perhaps just under-practised. Before acting on a trend, ask who actually did the primary research, and treat any bold claim with no named, checkable source as a red flag. Build the habit, in your own team and in how you consume content like this one, of asking “so what does this actually mean for us” before “how do we get on this.” It’s a small discipline, but it’s increasingly the difference between strategy and noise.
+ + + + + + + +
Are you thinking what we’re thinking, that keeping on top of how the future of work is evolving can be A LOT? Well, we live and breathe it, so you don’t have to, and we’re here to help.
Scroll down for more details on our future of work trend analysis, strategic foresight workshops and seminars, free webinars and consultancy services in this area.
+ + + + + + + +
2] The rise of the deepfake interviewee – hiring’s new trust problem
Here’s one that should give every hiring manager pause. A survey published in August 2026 by background-screening firm iprospectcheck, covering 1,500 US hiring managers and business owners, found that nearly a quarter, 22.9%, believe they may have unknowingly interviewed a proxy or AI-deepfake candidate: someone other than who they thought they were speaking to, or a real face paired with AI-generated answers.
This isn’t a one-off scam story. It’s a structural crack opening up in remote hiring itself. As video interviewing becomes the default and AI tools became good enough to convincingly fake a face, a voice or a set of answers in real time, the basic assumption behind a screening interview, that you’re actually talking to the person on the CV, has quietly stopped being safe to make.
For any organisation hiring remotely, and that’s most of us now, this is a trust and security problem hiding inside an HR process. A bad hire is expensive. A hire who isn’t who they claimed to be, in a role with access to systems, data or clients, is a different order of risk altogether.
Expect a fast-forming market of identity verification tools sitting between candidates and employers. In the meantime, it’s worth rebuilding at least one stage of your hiring process around something AI can’t easily fake: a live, unscripted task, a same-room or same-call verification step, or simply a harder question than “tell me about a time when.” Trust, in hiring as everywhere else, now has to be actively checked for, not assumed.
3] No right to be cool – the law that still doesn’t exist
Despite yet another run of UK record heatwaves this year, there is still no maximum legal workplace temperature in Britain.
It’s one of those facts that surprises almost everyone the first time they hear it: the Health and Safety Executive’s own position is that “no meaningful upper limit can be imposed”, since high indoor temperatures are often caused by the work itself rather than just the weather. Current law only requires a “reasonable” temperature, with guidance setting a minimum of 16°C, or 13°C for physically demanding work, but no ceiling at all.
That’s now becoming a live political and legal fault line rather than a quirk of employment law. In June 2026, the TUC called on employers to start cutting temperatures above 24°C and to let staff stop work above 30°C, or 27°C for strenuous roles, whilst the government’s own Climate Change Committee separately urged ministers to legislate a statutory maximum. So far, government has committed only to “modernising guidance”, not introducing an actual legal limit.
With no legal ceiling, where you draw the line on heat is entirely your own judgement call, not a compliance box-tick, and as heatwaves shift from exception to norm, “no law says we have to” is a weak position to hold with your own people, let alone with customers or investors watching your wider environmental commitments. Offices, warehouses, delivery fleets and outdoor teams all feel this unevenly.
Don’t wait for legislation to catch up. Set your own internal heat thresholds now. The TUC’s 24°C and 30°C bands are a sensible starting point, build them into your existing health and safety risk assessments, and communicate them clearly before the next heatwave hits rather than during it. Flexible hours and proper rest and hydration breaks cost far less than the reputational and productivity hit of being caught unprepared.
4] The advent of the infinite workday – and what that's doing to attention
According to Microsoft’s most recent Work Trend Index, the average employee is now interrupted every two minutes during core working hours, roughly 275 times a day, by meetings, emails and chats. Nearly a third of workers are back checking email by 10pm, and meetings held after 8pm are up 16% year on year.
Perhaps most telling of all, almost half of employees, 48%, and more than half of leaders, 52%, say their work now feels chaotic and fragmented. This isn’t a story about people working longer hours in the traditional sense. It’s a story about the working day losing its edges entirely, bleeding into evenings, weekends and every gap in between, whilst never actually feeling more productive for it.
Attention, not time, is becoming the scarcest resource in the modern workplace. A team that’s permanently interrupted cannot do the deep, careful thinking that good strategy and good judgement require, however many hours are logged on the clock. Left unmanaged, the infinite workday quietly erodes the very focus your business needs most, whilst looking, on paper, like commitment.
Treat focus time as seriously as you’d treat any other business-critical resource. Build meeting-free blocks into the working day, question whether every notification genuinely needs to interrupt someone in real time, and model switching off yourself, visibly and often. Leaders who protect attention will out-think leaders who simply out-work.
+ + + + + + + +
Business transformation isn’t the latest software or project tool. Lasting organisational change happens conversation by conversation...
So, if you’d like to explore anything we've touched on in this blog or discuss any other aspects of the future of work, please do get in touch.
You might also want to: