The Fog Machine: AI, Ambiguity, and the Quiet Push Out

Simon Carver

Welcome back to The Human Workforce. I'm Simon Carver, here with Lachlan Reed, and today we're talking about a pattern a lot of people can feel before they can name it. It's that strange corporate weather system where expectations are blurry, instructions are mostly verbal, and somehow the paper trail stays, very clean.

Lachlan Reed

Yeah. It's like being told to ride a trail you've never seen, in the dark, and then getting marked down for not taking the perfect line. Bit rough, mate. You hear things like, "Be more strategic," or "show more leadership presence," or "move faster with AI," but no one's telling you what good actually looks like in plain English.

Simon Carver

Right. And that's the part that matters. Ambiguity is not always accidental. Sometimes it's just disorganization. We should say that. Not every messy company is running a scheme. But sometimes ambiguity becomes a tool. If goals are vague, timelines shift, feedback stays verbal, and standards change after the fact, a worker can be kept in a permanent state of almost-but-not-quite good enough.

Lachlan Reed

The old "needs improvement" loop. Brutal little hamster wheel, that one. You ask, "What exactly needs improving?" and you get fog back. "Stakeholder alignment." "Executive presence." "Sharper thinking." Cheers, that clears up absolutely nothing. Even a kangaroo could trip over that.

Simon Carver

And emotionally, it wears people down. I think that's important to say out loud. This isn't just a process problem. It's a human problem. People start doubting their own memory, their own judgment, even their own track record. They think, maybe I really am slipping. Maybe I missed something obvious. But if the standard is never stable, confusion is the point, not the side effect.

Lachlan Reed

Yep. And then in rolls the shiny language. "Transformation." "Modernization." "Efficiency." "AI-first." Some of that is real. Some firms are genuinely redesigning work, fixing clunky systems, automating junk tasks, making life easier. Good on 'em. But sometimes, if we're honest, it's more like a fresh coat of paint slapped on a headcount reduction plan.

Simon Carver

Exactly. Real productivity improvement usually has fingerprints. Clearer workflows. Better handoffs. Less duplicate work. Faster cycle times. Fewer errors. Better service. Actual redesign. Symbolic AI rhetoric sounds different. It often starts with broad claims and ends with fewer people, while the underlying work still exists. It just gets redistributed, hidden, or deferred.

Lachlan Reed

And we've seen enough reporting, especially across 2025 and into early 2026, to know some companies appear to be making cuts in anticipation of AI gains, not always because the gains are already proven. That's the sneaky bit. The promise of future efficiency gets used like it's current fact.

Simon Carver

So if you're listening and thinking, this sounds familiar, you're not imagining the texture of it. The fog is real. The written deniability can be real. And when those things combine, workers can end up judged by standards that were never fully stated, in a game that was never meant to feel fair.

Lachlan Reed

So let's talk about who's most exposed here. Older workers, more experienced workers, long-tenured people... they can end up in the splash zone. Not because they suddenly forgot how to do the job, but because they often cost more, know where the skeletons are buried, and ask the annoying-but-important questions.

Simon Carver

That's such a key point. Experience brings institutional memory. It brings pattern recognition. It brings the ability to say, "We've tried this before," or "That metric doesn't measure what you think it measures," or "This vendor deck is overselling the result." In healthy organizations, that's valuable. In insecure ones, it can make a person inconvenient.

Lachlan Reed

Inconvenient is the word. If leadership wants compliance wrapped in enthusiasm, the person who calmly says, "Hang on, this doesn't stack up," can suddenly be framed as not agile, not modern, not aligned. It's the corporate version of blaming the smoke alarm for the fire.

Simon Carver

That's a very good analogy. And again, we need to be careful and grounded here. We're not saying every performance concern is fake. Some people do struggle. Some jobs really do change. Some automation is legitimate and overdue. But there is a recognizable pattern where shifting goals, undocumented feedback, and highly subjective reviews increase pressure on workers while lowering legal and reputational risk for the employer.

Lachlan Reed

Yeah, because if the direction's mostly verbal, and the standard's kinda floating around like a loose tarp in the wind, it's harder to point to one clean event and say, "There. That's the issue." Instead you've got this mushy story: concerns about fit, concerns about trajectory, concerns about adaptability. Not much to grab onto.

Simon Carver

And that mushiness matters. It can create what feels like a quiet exit ramp. No dramatic confrontation. Just repeated ambiguity, lowered ratings, shrinking scope, fewer invitations, less sponsorship, and subtle signals that maybe this is the time to step aside, retire early, or accept something diminished.

Lachlan Reed

Meanwhile the AI banner keeps flapping overhead. "We're transforming." "We're streamlining." "We're becoming leaner and more innovative." Maybe. Or maybe payroll is being cut and the story sounds nicer with machine learning sprinkled on top.

Simon Carver

That's the distinction we want to keep making. Real productivity improvement changes the work itself. It removes waste, clarifies decisions, automates repetitive steps, and gives people better tools. Labor cost reduction disguised as innovation mostly changes the budget line while leaving confusion in place. If the process is still broken, but there are fewer people and more slogans, that is not transformation. That's a smaller team standing in the same mess.

Lachlan Reed

Beautifully put. And experienced workers can see that difference faster than most. That's partly why they matter, and partly why they can become targets. They know what good operations look like. They know when "AI strategy" is real, and when it's just corporate oatmeal with a robot label on the tub.

Simon Carver

So the takeaway here isn't bitterness. It's clarity. If you're an experienced worker, your value may be exactly what the system is discounting: judgment, memory, ethics, context, and the courage to notice nonsense. The question is how to protect that value and make it legible.

Simon Carver

And that's where we pivot from diagnosis to tactics. Because the answer can't just be, "Well, good luck in the fog." Workers need tools. Used ethically, generative AI can help create clarity where the organization is withholding it.

Lachlan Reed

Yep. Think of AI as armour, leverage, and a bit of a megaphone. First move: after a meeting full of vague waffle, use AI to turn your notes into a crisp written follow-up. Something like, "Just confirming my understanding: the priority for Q2 is X, success will be measured by Y, and my next three deliverables are A, B, and C." Then send it. Calm, helpful, professional. Now the fog has to argue with a document.

Simon Carver

That alone can change the power dynamic. You're not accusing anyone. You're creating shared reality. AI can also help convert fuzzy objectives into measurable action plans. If a manager says, "Be more strategic," you can ask an AI tool to generate possible interpretations: stakeholder mapping, decision memos, risk flags, quarterly priorities, metrics. Then you choose the sensible ones and propose them. You are making ambiguity expensive.

Lachlan Reed

And it helps with plain old output too. Better writing. Cleaner presentations. Tighter status reports. Stronger executive summaries. If English isn't your first language, or if you're just knackered after a long day, AI can help polish without changing your thinking. That's not cheating. That's using a power tool instead of a rusty spanner.

Simon Carver

Another big one is proof of value. Use AI to build a personal skill portfolio: projects delivered, processes improved, revenue supported, risks avoided, teams mentored, systems understood. Ask it to help organize examples by theme and impact. Experienced workers often undersell themselves because so much of their value is invisible or relational. AI can help surface it.

Lachlan Reed

Also: process waste. This is a beauty. Feed an AI tool a rough map of your workflow and ask, "Where are the handoff delays, duplicate approvals, manual copy-paste steps, rework loops?" Then take the decent ideas and turn them into proposals. Real ones. If the company wants innovation, sweet as—show them actual automation that improves the work, not just cuts names off an org chart.

Simon Carver

And if the internal environment is deteriorating, AI can support mobility. Internal interviews. External job searches. Resume rewrites. Practice questions. Consulting offers. Even the early scaffolding for entrepreneurship: service descriptions, client proposals, workshop outlines, market research summaries. Not fantasy. Practical preparation.

Lachlan Reed

This is the bit I really want people to hear: don't let AI exist only as management's talking point. Use it yourself. Learn enough to be dangerous in the best sense. Dangerous to bad process. Dangerous to vague criticism. Dangerous to the idea that your value can't be measured or moved.

Simon Carver

Because the future probably does belong to workers who combine judgment, courage, communication, adaptability, and AI fluency. The deeply human strengths still matter—ethics, trust, mentorship, institutional memory, the ability to see patterns and call things by their proper name. AI can amplify those strengths. It does not erase them.

Lachlan Reed

So no surrender, no panic, no marinating in resentment. Get sharper. Get more documented. Get more portable. Get more human, and more AI-capable than the system expected. That's how you get harder to sideline.

Simon Carver

Well said. We'll keep building on this in future episodes. Thanks for spending time with us.

Lachlan Reed

Catch you next time, Simon.

Simon Carver

See you soon, Lachlan. Bye everyone.