Four Fire Alarms, No Exit

A friend asked me last week what I thought the biggest story in the world was right now. I gave her three answers before she’d finished her coffee, and none of them were the same story twice. That’s not me being scattered. That’s the actual problem.

We like our crises singular. One villain, one battlefield, one headline that can carry a week of coverage before something else eats it. Ukraine had a front line you could point to on a map. A pandemic had a curve you could watch flatten or spike. Even a stock market crash has a bell you can ring on a trading floor and a number that goes red. These are stories with shape. Journalists love them, audiences love them, and crucially, our brains are built to process them, because a single identifiable threat is something you can prepare for, fight, or flee.

What we’re apparently rubbish at is the other kind. The kind with no single villain, no single battlefield, and no bell. The kind that’s just… running, quietly, in five or six places at once, none of them dramatic enough on their own to lead the evening news, all of them compounding.

That’s where we are in the middle of 2026. Not one crisis. A stack of them, ticking along in parallel, mostly invisible to anyone who isn’t specifically paid to watch one of the individual dials.

The plumbing, briefly revisited

I’ve already written about the AI-cyber side of this at length in The Plumbing Problem, so I won’t re-tread the whole argument here. The short version, for anyone who missed it: the IMF spent a blog post in May explaining, in the driest institutional language available, that AI-accelerated cyberattacks are now a genuine financial stability risk rather than a niche IT problem. A single exploited weakness in AI or cloud infrastructure could ripple across many institutions at once, given how much of the financial system leans on the same small stack of shared software and cloud providers. The Fund pointed specifically at Anthropic’s Mythos preview as proof of concept… a model capable of identifying and exploiting weaknesses across every major operating system and browser, even in the hands of someone with no particular expertise.

That’s not theatre. Nobody’s uploading their consciousness or building a paperclip factory. It’s a much duller, much more plausible scenario: a vulnerability gets found faster than anyone can patch it, and the whole plumbing of the financial system, the bit nobody thinks about until it backs up, goes wrong all at once. I gave that story 2,000 words on its own, because it deserved them. Today it’s one line item on a longer list, which tells you something about the scale of what’s actually queuing up behind it.

The bigger picture nobody’s holding

Here’s where it gets genuinely uncomfortable. The World Economic Forum published its Global Risks Report for 2026 in January, and the top line isn’t cyber at all. It’s what they’re calling an “age of competition”… geoeconomic confrontation, the weaponisation of sanctions and trade and investment restrictions, sitting at the very top of the list for the first time, having jumped eight places in a single year. Armed conflict comes second. Extreme weather, societal polarisation, and misinformation round out the top five.

Half the people surveyed, and these are risk experts and business leaders, not doom-scrolling members of the public, expect the next two years to be turbulent or outright stormy. That figure rose fourteen percentage points in a single year. Stretch the horizon to a decade and it climbs further still, with the overwhelming majority expecting instability rather than calm.

What I find properly bleak about the report isn’t any single risk on the list. It’s the connective tissue. Inequality was named the most interconnected risk in the entire survey for the second year running, closely followed by economic downturn, with misinformation sitting just behind those. In plain English: these things aren’t sitting in separate boxes. They’re feeding each other. A weakened power grid is more vulnerable to a cyberattack. A cyberattack on a port amplifies a supply chain shock. A supply chain shock feeds inflation, which feeds political anger, which feeds the exact polarisation that’s already sat in the top five. Nobody designed this as a system. It just behaves like one anyway.

Zurich’s group chief risk officer, commenting alongside the WEF release, made a point I keep coming back to: the risks that actually erode a society over time, the slow ones, are chronically underweighted against the loud ones. Disruptions to critical infrastructure, despite touching energy, water, and digital systems all at once, ranked a startling twenty-third when people were asked to think ten years out. We are, in other words, extremely good at worrying about the thing currently on fire and remarkably bad at worrying about the thing that’s quietly rusting.

And AI threads through nearly all of it, not as a single discrete risk but as an accelerant sitting underneath several others at once. The WEF’s own analysis notes that the adverse effects of AI showed the single largest rise of any risk when you compare the two-year outlook to the ten-year one, climbing from thirtieth place to fifth. Separately, Allianz’s own annual risk survey, a different methodology entirely, run on businesses rather than policy experts, landed on almost the same conclusion from a different angle: cyber incidents topped their list for a fifth consecutive year, with AI as the fastest-rising concern, climbing from barely inside the top ten to second place in twelve months. Two entirely separate surveys, different populations, different questions, and they both point the same direction. That’s not noise. That’s a pattern.

Why this doesn’t make the news, and the news that does

None of this travels well. That’s the actual crux of it. A war has soldiers, a front, casualty figures you can update daily. A flood has footage. Even a scandal has a villain you can put a face to and a headline you can write in eight words.

What do you do with “the interconnectedness of global systemic risk is increasing”? You can’t put that on the ten o’clock news with a dramatic sting. You can’t march against it. There’s no single meeting where world leaders sign a document and it stops. It just… continues, in the background, distributed across a hundred different technical reports that ninety-nine percent of people will never open, each one perfectly reasonable on its own and terrifying in aggregate.

So instead we get fragments. A headline about a bank outage here. A paragraph about a ransomware attack on a hospital there. A short segment on AI job displacement, followed immediately by something about the football. Each fragment lands, gets its ninety seconds of attention, and disappears, never once connected to the fragment before it or the one that’ll follow next week. Nobody’s lying to you exactly. Nobody’s hiding the plumbing. It’s just that the plumbing is boring, distributed, and technical, and boring distributed technical things have never once won a ratings war against a single dramatic story with a face attached.

I say this from a cave in rural Spain, which is either the worst possible vantage point for commentary on global systemic risk or, honestly, the best one. Out here the wifi drops when it rains, the water comes from a well half the year, and my sense of “infrastructure” is a good deal more literal than most people’s. You notice fragility faster when you’re occasionally living without the thing that’s meant to be fragile. It sharpens the cynicism rather than dulling it.

The actual unsettling conclusion

So here’s where I land, and it’s not a comforting place. The most disturbing thing happening right now might not be any individual item on any of these lists. It might be the structural fact that there are now multiple serious, compounding, technically sophisticated problems running concurrently, each one alone would have been a defining story in a quieter decade, and the sheer volume of them is what’s actually keeping any one of them from getting the attention it deserves.

We’re not facing one crisis that everyone can rally around or against. We’re facing a dozen half-visible ones, each competing for the same shrinking pool of public attention, each one slightly too technical, too slow, or too distributed to win. The IMF worries about the plumbing. The WEF worries about the geopolitics. Zurich worries about the infrastructure nobody’s maintaining. Allianz worries about the businesses caught in the middle. They’re all, in their own dry institutional way, describing the same thing from different rooms in the same house. And the house is the one we’re all still living in, mostly unaware that four separate fire alarms are going off in four separate corridors, none of them loud enough on its own to make anyone actually leave the building.

That’s not a call to panic, because panic is just another fragment, another ninety-second segment that burns bright and then gets replaced. It’s a call to notice the pattern rather than the individual sparks. The plumbing, the geopolitics, the infrastructure, the inequality, the misinformation… none of them are new stories exactly. What’s new is how many of them are running at full speed, at the same time, largely unwatched, because we’re all still waiting for someone to hand us a single villain to point at.

There isn’t one. That’s rather the point.

Until Next Time