Business locked in costly AI 'arms race'

There is no doubt that we are in an AI arms race, says Jon Collins.

He worked in IT for 35 years in various roles, including as a software programmer, system manager and chief technology officer.

He is now an industry analyst for research firm Gigaom.

Mr Collins says the current arms race was spurred by the launch of the Chat GPT in late 2022.

Since then, many such generative AI systems have emerged, and millions of people use them every day to create artwork, text or video.

The stakes are high for business leaders. Generative AI systems are powerful tools that can digest more data in minutes than a human can in several lifetimes.

Suddenly company leaders are realizing what AI can allow them and their competition to achieve, Mr. Collins explained.

“Fear and greed is driving it,” he says. “And that creates an avalanche of momentum.”

With the right training, a custom AI system can allow a company to outpace its competitors with research breakthroughs, or reduce costs by automating tasks currently performed by humans.

In the pharmaceutical sector, firms are customizing AI to discover new compounds to treat disease. But this is an expensive process.

“You need data scientists, and you need model engineers,” Mr. Collins explains.

These scientists and engineers need to have at least some understanding of the pharmaceutical sector in which AI will work.

And it doesn't stop there. “You need infrastructure engineers who can build your AI platforms,” ​​he continues.

Such highly skilled workers are not easy to come by.

“Understanding how to build these systems, how to make them actually perform, and how to solve some of the challenges ahead,” says Andrew Rogowski, director of innovation at the Surrey Institute for People-Centered AI. There aren't enough people.” at the University of Surrey

He added that the salaries of those who can handle these challenges have risen to “ridiculous” levels, because they are so important.

“If we had the capacity, we could produce hundreds of AIPhDs, because people would give them jobs.”

In addition to the lack of expertise, gaining access to the physical infrastructure required for large-scale AI can be a challenge.

The kind of computer system needed to run AI for cancer drug research typically requires between two and three thousand sophisticated computer chips.

The cost of such computer hardware can easily top $60m (£48m), even before the cost of other essential equipment such as data storage and networking.

Part of the problem for businesses is that this type of AI has appeared suddenly. Previous technologies, such as the emergence of the Internet, were developed more slowly.

A large bank, pharmaceutical firm or manufacturer may have the resources to buy into the tech needed to take advantage of the latest AI, but what about a smaller firm?

Italian start-up Rest World is a recruitment website for catering staff, with a database of 100,000 workers.

Chief Technology Officer Eduardo Conte wanted to see if AI could benefit business.

The firm considered building an AI-powered chatbot to interact with users of the service.

But Mr. Conte said that, among thousands of users, “the cost goes up a lot.”

Instead, he looked at a narrower problem — the problem that candidates don't always best represent their experience.

For example, a candidate cannot list hospitality as a skill. But the algorithms Mr Conte has developed make it easier to uncover additional information, including whether he has applied in the past and won a waiting role.

“The AI ​​can infer that they're a waiter, or maybe they're interested in other waiter jobs,” he says.

One of the hurdles in hospitality recruitment is getting candidates to the interview stage.

So, Mr. Conte's next challenge is to use AI to automate and customize the interview process for his candidates.

AI can also “convers” with candidates and generate summaries to deliver to recruiters.

This could speed up the entire process, which can currently take a few days, while a waiter or chef could find another job.

In the meantime, big firms will continue to pour cash into AI projects, even if it's not always clear what they're likely to achieve.

As Mr. Rogowski says, AI adoption is in a “Darwinian, experimental phase” and it's hard to see what the consequences will be.

“That's where it gets interesting. But I think we have to go with it,” he says, “I'm not sure we'll have a choice.”