There’s an unwritten rule in publishing, or so I’ve been told: don’t write about COVID. Our collective attention span has been saturated by those endless months holed up in attics and cramped corners of apartments, staring out at a world we could no longer take part in. When the worst of it passed, we felt an urge to close that chapter, to padlock it behind a heavy latch.
But in doing so, we also tuck away the hard-won lessons of that time: how quickly systems buckle, how two decades of coronavirus warnings accumulated without adequate preparedness, and how the very mechanisms we rely on for safety can become the scaffolding of a next disaster.
This matters now as another threat is taking shape: highly pathogenic avian influenza, known as bird flu.
Bird flu still poses a low‑probability threat of sustained human transmission. But that doesn’t make the virus harmless. The H5 viruses are brutally lethal to birds – 9 million have died outright, and hundreds of millions have been culled to contain the spread. Alarming is the virus’s expanding reach into mammals. So far, at least 74 mammal species, from elephant seals to polar bears, have suffered die‑offs.
The individual cases are situated within a broader shift. Dense poultry farms create opportunities for the virus to hop species. Over a thousand US dairy herds have tested positive in the past two years, and viral fragments have even been detected in milk – a worrying route of spillover. Every jump is a probe for new footholds.
Europe is seeing a surge too. From early September to mid-November 2025, 1,444 infected wild birds were found across 26 countries: a quadrupling compared with the year before.
Understand how AI is changing society, with our weekly newsletterSubscribe for free
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Human cases remain rare: only 992 confirmed H5N1 infections worldwide since 2003, though with a near‑50% fatality rate. But the numbers are increasing.
The Americas have logged 75 cases since 2022, and in November, the US recorded its first H5N5 death in a patient with existing health problems. And although no human cases have been reported in Europe, the European Centre for Disease Prevention and Control warns that the widespread animal circulation raises the risk of spillover.
My research focuses on how warnings collapse before catastrophe, from geopolitical shocks to intelligence failures and industrial accidents. The pattern is often the same. Frontline observers spot something early, but the signal fades as it moves upward, diluted by bureaucracy, competing interpretations, or institutional forgetfulness.
The recent Hong Kong fire is yet another tragic example: residents at Wang Fuk Court had raised multiple alarms about the styrofoam boards that ignited with a lighter, the uncertified netting and the pattern of ignored safety notices long before the blaze, yet those concerns never gained traction.
The failures I study share recurring blind spots: weak signals drowned out by noise, bureaucratic habits that slow or soften uncomfortable messages, and the political instinct to downplay problems that threaten established narratives. When you see warning as a chain running from detection to decision, collapse is often partial. Some links hold. Others jam at the moment they are most needed.
Bird flu now sits inside that kind of chain. The technical ability to detect change is there: veterinarians, virologists and surveillance systems are picking up signals, sequencing viruses and logging outbreaks. But the infrastructure meant to catch the virus in its early stages is fraying. The agencies that once charted the terrain of emerging pandemic threats have been hollowed out – budgets trimmed, staff evaporated.
Surveillance falters
A study of 31 European nations warned that COVID exposed a “critical gap in preparedness” and urged standardised indicators and open data as the foundation for any future response. The EU’s freshly launched pre-pandemic plan is a good step, but it cannot mask the gaps in day-to-day monitoring and response that still leave countries exposed.
Across the Atlantic, cuts have left the US Centers for Disease Control and Prevention scrambling. American scientists warn that federal reporting has slowed: the United States Department of Agriculture shared too little genetic data on the outbreak in cattle and other affected animals, released it late, and in formats researchers could not use. It left scientists unable to track how the virus was evolving or spreading across herds.
In the UK, domestic surveillance capacity has equally faced strain, with reduced access to European disease intelligence and chronic vet shortages weakening early detection.
Once the signal dims within institutions, it dims for the public as well. And a weak warning rarely travels far.
A recent poll shows this clearly: most Americans don’t even register bird flu as a credible threat. What doesn’t help is that symptoms in humans can be so mild that they slide past notice. A case in a dairy worker earlier this year looked like nothing more than conjunctivitis.
None of this means a new pandemic is imminent. Health authorities still say the chance of an efficient human-to-human outbreak is low. These viruses rarely make that leap. And we’re not helpless. We’re better prepared than we were before COVID: we have vaccine candidates, clearer protocols and agencies that learned painful lessons.
But low isn’t none. And if it were to occur, the consequences could be catastrophic. Most people have some immunity to the seasonal flu strains. We probably have none to H5.
And influenza doesn’t restrict itself to the frail in the way COVID often did; past flu epidemics killed healthy adults in large numbers. Adding to the concern, health expertise itself has come under attack, weakening the very authority that should turn signals into action.
If we avert our eyes from the bird flu threat because our systems have grown inattentive, underfunded and unprepared, we risk repeating that same pattern. And the next alarm will arrive too late for anyone to claim they didn’t see it coming.
