NEW AI BOOK IN BESTSELLER LIST FOR 12 MONTHs IN AIRPORTS GLOBALLY - WH SMITHS

AI will create new efficiencies for insurance. The benefits of AI to insurers will be huge and will result in lower premiums and greater profit:

◆ More accurate, instant underwriting/quotes based on smarter prediction of risks

◆ Linking of premiums to live data, e.g. miles driven by younger drivers

◆ More accurate grouping of risks for re-insurance with larger companies

◆ Faster and more efficient claims handling

◆ Better fraud detection

NEW AI BOOK IN BESTSELLER LIST FOR 12 MONTHs IN AIRPORTS GLOBALLY - WH SMITHS

AI SPENDING BY BANKS is growing by 30% a year and is likely to reach $100bn by 2027, mainly in North America. Last year, banks in the US and Canada were responsible for 80% of published research on AI and funded 60% of all banking investments in AI globally. While most of this is about saving on costs, as we will see, the greatest risk to banks will be AI-driven cyberattacks.

The banking sector as a whole is already in trouble. Indeed, it never recovered its full strength after the 2008–2009 banking crisis. A report by McKinsey shows that 75% of banks are delivering very low returns on capital for investors, not even covering the costs of providing that capital.

One reason is much tougher regulation, which hit profits by requiring banks to tie up more capital in reserves and be more cautious. But if you don’t allow decent financial returns on investment in a particular sector, then the market will tend to dump those shares in preference for something ‘safer’. And this is what has happened: banks are now struggling to attract strong investor interest – compared to ‘more exciting’ AI-related companies for example. But our world needs strongly capitalised banks. Banks are as fundamental to the future of every country as hospitals and schools.

NEW AI BOOK IN BESTSELLER LIST FOR 12 MONTHs IN AIRPORTS GLOBALLY - WH SMITHS

AI WILL CUT banking costs, improve decision-making and reduce certain risks, in particular from fraud. But AI decisions will also dominate global markets, destroy some banks and risk global economic chaos.

A very important area of financial services is wealth management, and much of this relates to pensions, because for most people almost all the wealth they have, other than their homes if they are owners, is in pension savings (unless they work for government).

AI is taking control of stock markets globally

AI is rapidly taking control of all global stock markets. More than $110tn of company shares are listed in these exchanges, but over 90% of trading decisions are by computers, nanoseconds after a small or large event. Human beings are no longer involved. And increasingly such decisions are controlled by AI.

In the past, computer trades were automated, using human rules linked to market data, social media and news bulletins. The key to ‘algorithmic trading’ is ultra-speed. If there’s an earthquake in San Francisco, or an assassination in Brazil, computers dump huge volumes of shares a few billionths of a second later. That’s because their rules are set to ‘expect’ many others to ‘panic sell’.

The most successful trading companies make tens of millions of dollars a year by sending in their own automated orders just a few billionths of a second before anyone else. They have faster processors, shorter cables, and use other digital tricks such as their own microwave transmitters to avoid latency in public networks.

NEW AI BOOK IN BESTSELLER LIST FOR 12 MONTHs IN AIRPORTS GLOBALLY - WH SMITHS

MANUFACTURERS ARE ALREADY using AI to help reduce the impact of production on the environment, but AI itself is creating a new hazard by using vast amounts of additional energy and water (for cooling).

The digital world is already responsible for around 4% of global carbon emissions, more than the entire aviation industry, with 4.6 billion people connected in some way to 30 billion devices. Even before AI, energy consumption by data centres was growing by 25% every year, despite the fact that servers are now 20% more efficient than five years ago. 

AI is going to consume vastly more power. Gartner predicts that AI alone could burn up a further 3.5% of the world’s energy by 2030, almost doubling the environmental footprint of today’s digital world.

By 2026, AI servers will need as much power as the whole of Argentina, the Netherlands or Sweden – 130 terawatts per year or 0.5% of entire global energy consumption.