The turning point for artificial intelligence, which may or may not be dominated by the major cloud providers

Artificial intelligence and machine learning require huge data processing and storage capacity, making the cloud the preferred choice. That raises the specter of some of the cloud giants dominating AI applications and platforms. Can the tech giants control the AI ​​narrative and reduce the choice for businesses?

Not necessarily, but with some caveats, emphasized AI experts. But the big cloud providers certainly have control over the AI ​​story from many perspectives.

That was part of the consensus raised at a recent webcast by New York University Center for the Management of the Future and LMU . Institute of Strategy, Technology and Organizationjoined by Daron Acemogluprofessor at MIT; Jacques Bughin, professor at Solvay School of Economics and Management; and Raffaella Sadunprofessor at Harvard Business School.

There’s more to AI than the cloud. The complexity and variety of AI applications extend beyond the cloud environments in which they are run – and thus diminish the dominance of a few cloud giants.

Undoubtedly, “AI will require more space in storage, information flow,” Bughin said. At the same time, “the cloud is just one part of the total platform pie. It’s part of the infrastructure, but the foundation layer is something that you develop in-house and through third parties. This integration is going to be the future, even more important than the cloud itself.Let’s be very clear, it’s not operations, it’s lots of algorithms, lots of different data, that integration, will be. requires system integration, architecture and design. That means different types of companies will be involved in that work.”

What Bughin is more worried about is that the innovation potential from AI startups could be stifled by larger players eating up smaller companies and startups through mergers and acquisitions. “Companies like the big Internet or those AI companies are going and buying a lot of very small and very smart AI companies.”

At the same time, Sadun points out, smaller companies may be in a better position to capitalize on AI innovations — but need help with training and education to prepare for them. “The issue of who benefits from AI is really important,” she said. “On the one hand, we might think that smaller companies can use these technologies more effectively, because they’re more agile, more flexible. Companies that already have digital can exploit it.” and scale AI.”

Where the big cloud providers can also feel their dominance is monopolizing the data that feeds AI systems, Acemoglu said. The cloud architecture itself can be based on competitive and price-sensitive cloud services, he explained. “But the cloud architecture won’t let you mine the data. The areas where I worry about the future of AI technology are the areas that allow companies to monopolize the data. That’s where the companies are. company has too much influence over the future direction of technology, which means that a few people in the boardroom will determine where technology is. We want more people to be focused and people-centered That wouldn’t be possible if a few companies had different business models that dominated the future of technology.”

Bughin believes the value of an AI-based business “is not in the cloud that allows it to work”. “I think there’s enough competition that the price doesn’t destroy the value. The value will come from the fact that you’ve integrated these technologies into where you work and how your company operates, on the back end. your own. won’t be the battlefield. Value comes from generating productivity and revenue, at a faster rate than we’ve seen in traditional digital transformations.”

And, for the first time, we see the terms traditional and digital transformation used together in the same sentence. As these thought leaders are concerned, such transformations are moving to the next stage, enabling automated, software-driven operations and innovation through AI. It’s a question of whether the big tech vendors can control the momentum or if it remains a market and practice with a diversity of options. Keep stable.