Artificial intelligence seems to be taking over the world these days — what with repeated scary predictions in the press and social media.

These days, up to 10 per cent of the articles in science and medicine journals contain some aspect of AI.

What AI is really good at is handling and identifying patterns in mounds of data to identify, for example, a specific person and even something of their mood from their gestures, voice and face.

Or, in the world of biochemistry, AI is proving to be a powerful tool for identifying patterns in the sequence of hundreds, or even thousands, of amino acids in proteins and hence the shapes and functions of those proteins.

Given that those proteins were created from DNA and RNA blueprints, mutant versions of either can cause trouble by creating proteins better able to evade the body’s natural defenses or drugs in the case of cancers, or bacterial and viral infections.

The more malignant the cancer, the more mutations are likely to be found. Cancer cells, especially the really malignant of them, are prime examples of evolution on the fast track – the more cell divisions, the more mutations, the greater the chance that one or more, or perhaps many mutants, will prove capable of overwhelming the body’s natural defences and any treatment.

That’s why some cancers, such as glioblastoma multiforme, the most malignant of all brain tumours, are next to impossible to treat.

The five-year survival rate is less than five years and hasn’t budged for many years. Genomic studies of this cancer exhibit different mutant versions of genes in different regions of the affected brain – different colonies of malignant cells working out their own evolutionary solutions for survival.

Mutations are a big problem with some bacterial infections that manage to mutate their way around antibiotics, or the COVID-19 virus, which has continued to mutate over the three to four years of the pandemic.

Most of COVID’s mutant versions cause no trouble, but some as we witnessed during the pandemic, conferred advantages to the virus by making some variants far more catchable, others more lethal and yet others find their way around protections offered by earlier vaccines.

And much to our chagrin, we’ve also learned that the protection offered by vaccines lasts only six or so months before the need for booster shots and even updated shots.

The problem is that COVID, like other viral infections, has billions of human hosts — to say nothing of nature’s other creatures — which can harbour the virus, which is free to mutate.

Updating mRNA vaccines has helped, but those versions are already old news and don’t take account of potential future mutant threats. That’s where AI comes in.

Given enough data about evolutionary changes in the COVID genome and spike protein collected over the last four years from around the world, it should be possible for AI to predict possible new variants before they take place and identify which ones are most likely to cause significant clinical outbreaks.

That would make it possible to design mRNA vaccines that are based not only on yesterday’s variants but possibly yet to develop future variants. That way, we stay ahead of the virus, not always a step behind.

Fortunately, unlike other vaccines, mRNA vaccines such as Moderna’s and Pfizer’s are much easier to modify and produce within short periods.

There is one major problem with the current mRNA vaccines – protection doesn’t last long. This was apparent as early as 2021 in Israel where the evidence suggested that booster shots were necessary in as little as six months following the first shot. The reluctance of so many to continue booster shots compounds this problem.

There is another solution. Develop a universal vaccine based on all prior variants of SARS-Cov 2 and selected closely related viruses including some in animals in close contact with humans.

The U.S. National Institutes of Health recommended this strategy twice before as did the now-retired Anthony Fauci, but it died for lack of funding and the pressing need in 2020 to get the mRNA vaccines as early as possible in the pandemic.

This year the Nobel committee awarded a Nobel Prize to two scientists who overcame repeated funding and other obstacles to put mRNA technology on a solid foundation, without which we would not have had two highly effective mRNA vaccines within a year of the start of the pandemic.

But the work isn’t over to create more effective, longer-lasting versions of the mRNA vaccines for COVID and possibly for the flu and even the common cold.

Remember to get your flu shot and updated COVID shot soon, and if you are 60 or older, consider getting the RSV shot, though for now, that requires a script and charge.