The Mexican swine flu, a genetic chimera probably conceived in the faecal mire of an industrial pigsty, suddenly threatens to give the whole world a fever. The initial outbreaks across North America reveal an infection already travelling at higher velocity than did the last official pandemic strain, the 1968 Hong Kong flu.

Stealing the limelight from our officially appointed assassin, H5N1, this porcine virus is a threat of unknown magnitude. It seems less lethal than Sars in 2003, but as an influenza it may be more durable than Sars. Given that domesticated seasonal type-A influenzas kill as many one million people a year, even a modest increment of virulence, especially if combined with high incidence, could produce carnage equivalent to a major war.

Meanwhile, one of its first victims has been the consoling faith, long preached by the World Health Organisation, that pandemics can be contained by the rapid responses of medical bureaucracies, independent of the quality of local public health. Since the initial H5N1 deaths in Hong Kong in 1997, the WHO, with the support of most national health services, has promoted a strategy focused on the identification and isolation of a pandemic strain within its local radius of outbreak, followed by a thorough dousing of the population with antivirals and (if available) vaccine.

An army of sceptics has contested this viral counter-insurgency approach, pointing out that microbes can now fly around the world (quite literally in the case of avian flu) faster than WHO or local officials can react to the original outbreak. They also pointed to the primitive, often non-existent surveillance of the interface between human and animal diseases. But the mythology of bold, preemptive (and cheap) intervention against avian flu has been invaluable to the cause of rich countries, like the US and UK, who prefer to invest in their own biological Maginot lines rather than dramatically increasing aid to epidemic frontlines overseas, as well as to big pharma, which has battled developing-world demands for the generic, public manufacture of critical antivirals like Roche's Tamiflu.

The swine flu may prove that the WHO/Centres for Disease Control version of pandemic preparedness - without massive new investment in surveillance, scientific and regulatory infrastructure, basic public health, and global access to lifeline drugs - belongs to the same class of Ponzified risk management as Madoff securities. It is not so much that the pandemic warning system has failed as it simply doesn't exist, even in North America and the EU.

Perhaps it is not surprising that Mexico lacks both capacity and political will to monitor livestock diseases, but the situation is hardly better north of the border, where surveillance is a failed patchwork of state jurisdictions, and corporate livestock producers treat health regulations with the same contempt with which they deal with workers and animals. Similarly, a decade of urgent warnings by scientists has failed to ensure the transfer of sophisticated viral assay technology to the countries in the direct path of likely pandemics. Mexico has world-famous disease experts, but it had to send swabs to a Winnipeg lab in order to ID the strain's genome. Almost a week was lost as a consequence.

But no one was less alert than the disease controllers in Atlanta. According to the Washington Post, the CDC did not learn about the outbreak until six days after Mexico had begun to impose emergency measures. There should be no excuses. The paradox of this swine flu panic is that, while totally unexpected, it was accurately predicted. Six years ago, Science dedicated a major story to evidence that "after years of stability, the North American swine flu virus has jumped onto an evolutionary fasttrack".

Since its identification during the Great Depression, H1N1 swine flu had only drifted slightly from its original genome. Then in 1998 a highly pathogenic strain began to decimate sows on a farm in North Carolina and new, more virulent versions began to appear almost yearly, including a variant of H1N1 that contained the internal genes of H3N2 (the other type-A flu circulating among humans).

Researchers interviewed by Science worried that one of these hybrids might become a human flu (both the 1957 and 1968 pandemics are believed to have originated from the mixing of bird and human viruses inside pigs), and urged the creation of an official surveillance system for swine flu: an admonition, of course, that went unheeded in a Washington prepared to throw away billions on bioterrorism fantasies.

But what caused this acceleration of swine flu evolution? Virologists have long believed that the intensive agricultural system of southern China is the principal engine of influenza mutation: both seasonal "drift" and episodic genomic "shift". But the corporate industrialisation of livestock production has broken China's natural monopoly on influenza evolution. Animal husbandry in recent decades has been transformed into something that more closely resembles the petrochemical industry than the happy family farm depicted in school readers.

In 1965, for instance, there were 53m US hogs on more than 1m farms; today, 65m hogs are concentrated in 65,000 facilities. This has been a transition from old-fashioned pig pens to vast excremental hells, containing tens of thousands of animals with weakened immune systems suffocating in heat and manure while exchanging pathogens at blinding velocity with their fellow inmates.

Last year a commission convened by the Pew Research Center issued a report on "industrial farm animal production" that underscored the acute danger that "the continual cycling of viruses ... in large herds or flocks [will] increase opportunities for the generation of novel virus through mutation or recombinant events that could result in more efficient human to human transmission." The commission also warned that promiscuous antibiotic use in hog factories (cheaper than humane environments) was sponsoring the rise of resistant staph infections, while sewage spills were producing outbreaks of E coli and pfiesteria (the protozoan that has killed 1bn fish in Carolina estuaries and made ill dozens of fishermen).

Any amelioration of this new pathogen ecology would have to confront the monstrous power of livestock conglomerates such as Smithfield Farms (pork and beef) and Tyson (chickens). The commission reported systemic obstruction of their investigation by corporations, including blatant threats to withhold funding from cooperative researchers .

This is a highly globalised industry with global political clout. Just as Bangkok-based chicken giant Charoen Pokphand was able to suppress enquiries into its role in the spread of bird flu in southeast Asia, so it is likely that the forensic epidemiology of the swine flu outbreak will pound its head against the corporate stonewall of the pork industry.

This is not to say that a smoking gun will never be found: there is already gossip in the Mexican press about an influenza epicentre around a huge Smithfield subsidiary in Veracruz state. But what matters more (especially given the continued threat of H5N1) is the larger configuration: the WHO's failed pandemic strategy, the further decline of world public health, the stranglehold of big pharma over lifeline medicines, and the planetary catastrophe of industrialised and ecologically unhinged livestock production.