Hinton, co-inventor of neural networks, receives Nobel Prize in Physics, fears AI

On October 8, the 2024 Nobel Prize in Physics went to a computer scientist and a physicist for “fundamental discoveries and inventions that enable machine learning with artificial neural networks,” according to the New York Times.

Since neural networks fall more into the field of computer science than physics, this message raised many questions in my mind:

• Couldn't the Academy award the prize to physicists who have discovered something new in physics?
• Should we join Hinton in fearing that his creation—“something smarter than us”—might have “bad consequences”?
• Will Hinton's Nobel Prize have any impact on how companies use AI?

The answers: Possibly; some already do; and no.

Nobel Prize in Physics 2024

The Royal Swedish Academy of Sciences awarded the 2024 Nobel Prize in Physics to two professors: John Hopfield, a physics professor at Princeton University, and Geoffrey Hinton, a computer scientist at the University of Toronto. The academy awarded the prize for its discoveries that helped computers “learn more like the human brain does,” the academy noted Just.

Although such neural networks represent a huge breakthrough, their connection to physics is not their main feature. To be fair, the Nobel Committee said that neural networks play an important role in scientific research – including in creating “new materials with specific properties.” Just wrote.

According to this logic, the Nobel Committee could have justified awarding the Prize for Physiology and Medicine. How come? Finally, among many other such models, researchers at Stanford Medicine have also developed SyntheMol, an AI model to create “recipes for chemists to synthesize drugs in the lab,” it says Stanford Medicine News Center.

The 2024 Nobel Prize in Physiology and Medicine went to two Massachusetts researchers – Victor Ambros and Gary Ruvkun – for their discovery of microRNA, “which helps determine how cells develop and function,” according to the statement Just. I wonder why the Nobel Committee didn't award the prize to researchers closer to the center of the physics bullseye.

Should we fear the dire consequences of AI?

We should fear the dire consequences of AI – and we already do.

Hinton spoke to journalists about the importance of neural networks. “It will be comparable to the Industrial Revolution,” he said, according to the Just. “Instead of surpassing humans in physical strength, it will surpass humans in intellectual ability. We have no experience of what it’s like to have things smarter than us.”

Although he believes technology will increase healthcare productivity, he worries about its downsides. He expressed concern “about a range of possible dire consequences, particularly the risk of these things spiraling out of control,” the noted Just.

This summer I sat in on a conversation with retail industry board members and witnessed their palpable horror. My interpretation is that these board members are torn between strong emotional poles.

The bright green light says companies should invest heavily in AI to avoid falling behind the competition, while the bright red light holds companies back because of the risk that AI hallucinations could potentially damage the company's reputation and lead to lawsuits.

Will this price impact how companies use AI?

This Nobel Prize will have no impact on how companies use AI because the hype has already peaked.

Meanwhile, these competing emotional poles are holding many companies back from offering generative AI to their customers.

How come? Of the 200 to 300 generative AI experiments that a typical large company runs, about 10 to 15 have resulted in widespread internal rollouts, and according to my June report, perhaps one or two are being released to customers Forbes Interview with Liran Hason, CEO of Aporia, a startup that sells companies a system that detects AI hallucinations.

In my conversations with business leaders since publishing my book in July Brain rushTwo questions from business leaders predominated:

• Which generative AI applications bring the most profit?
• What role should the CEO play in leading AI design and deployment?

Looking for profitable generative AI applications

I have a different idea of ​​what an AI payout would look like. Instead of focusing on how long it takes for a company's profits to cover the costs of building AI, I think business leaders should focus on a different metric.

What is particularly important for investors is that growth exceeds expectations. Therefore, business leaders should use generative AI to drive the development of new, fast-growing products and services that can offset the company's reliance on mature core products.

Only a few companies actually manage to do this. As I noticed in mine Pyramid of values Case Study: Most generative AI use cases help people overcome creator blocks, such as the fear of writing an email. Less generative AI applications help improve the productivity of business functions such as customer service or coding. And few, if any, applications of AI chatbots enable companies to create new revenue streams.

The role of the CEO in designing and deploying AI

One reason there are so few generative AI applications at the top of the value pyramid is the role of the CEO.

To understand why, consider two approaches to new product development: relay racing and rugby.

In the relay race approach, engineers develop a design for a new product. When they are finished, they hand the baton for the blueprint to the production manager, who protests that producing the blueprint would be too expensive and would lead to quality problems.

Once manufacturing has built what the engineers ordered, the manufacturing manager passes the baton to the sales manager and asks him to find customers for the products at the loading dock. The sales manager reviews the products and says they don't have the features customers want and are difficult to sell.

Now consider the rugby approach. Here, the CEO assembles a scrum made up of executives from sales, purchasing, manufacturing and finance. The Scrum visits early adopter customers and listens to them talk about their unmet needs.

Scrum then develops a prototype – taking into account the concerns of each business function and how well the new product will meet the needs of early adopters.

The Scrum then hands over the prototype to the customer and asks for feedback. Typically, this results in the team adding new features, deleting others, and after a few iterations producing a product that customers are eager to buy.

Last month I did a podcast with retail managers who told me that the typical retail CEO uses the relay race approach. When it comes to generative AI, the most common approach is similar: the CEO delegates design and deployment to a chief AI officer who lacks a deep understanding of business strategy.

Instead of delegating AI, the CEO needs to understand how AI can help reshape the way a company works with customers to create value, modeled on the Rugby approach described above.

A case in point is TechSee, a company that has developed a much more effective way to deliver remote customer service. According to my October report, company founder and CEO Eitan Cohen was unable to help a family member resolve a computer problem over the phone Inc. Split.

His mother-in-law called and said her printer wasn't working. “People can’t communicate the problem over the phone,” he told me.

“I had to drive over to look at the printer to figure out what was wrong and fix it. I didn't like that. I thought there must be a solution. I could take remote control of the computer. But that doesn't help me see that the dog chewed on the cable or that it wasn't plugged into the printer in the right place,” he added.

TechSee's solution pays off for customers. “We enable our customers to go from defense to offense,” he said. “Contact center agents can resolve customer issues quickly, freeing up time to sell products. We are reducing the average call duration by 20 to 50 percent – ​​from an hour to less than 30 minutes,” he added.

TechSee's solution allows customers to send images of the products they are seeking service for. These images reveal the root of the problem in a way that words cannot. Additionally, the company leverages its ability to analyze the most common issues and enable faster self-service problem resolution using an AI chatbot.

If more companies follow TechSee's approach to using generative AI, there could be more good that Hinton helped create and less of the bad.