Loading summary
Transcript22 lines
- [00:01]
AWS Narrator
How do more than 100 million Fortnite players join the battle without lag? AWS is how epic games scales up to keep them in the action. AWS powers next level innovation for millions of businesses.
- [00:19]
Julie Chang
Welcome to Tech News briefing. It's Tuesday, December 9th. I'm Julie Chang for the Wall Street Journal. AI agents are here debugging code, designing products and delivering ROI. We'll tell you how companies Walmart and BNY are already seeing results and what it could mean for the broader market. Plus why Russia is lagging behind in the AI supremacy race.
- [00:46]
But first, AI agents are tools that can perform actions on behalf of humans, such as booking a restaurant reservation or debugging code. While some critics question the value of this relatively new tech the the Wall Street Journal's chief of the Enterprise Technology Bureau, Stephen Rosenbush, spoke with our colleague Peter Ciampelli about the returns some businesses are already seeing.
- [01:09]
Peter Ciampelli
Stephen, some early adopters say that they're seeing the value of deploying AI agents already. For companies who are seeing that return, how are they using AI?
- [01:18]
Stephen Rosenbush
The handful of companies that are seeing a really meaningful return, in many cases they're using agents behind the scenes. They're using them for coding and customer service, or they're using them deep in the enterprise for, say, financial operations or supply chain management. And they tend to be applications where you have a lot of people or a lot of tasks that are very similar. They're not one of a kind applications or problems. Let's say you have accounting. The accounting principles don't change, or at least they don't change all that often. You try and have people applying those principles in a uniform way, and those kinds of problems can then lend themselves more readily to an AI agent. At one end of the spectrum, you can also have an agent that is functioning as an assistant to an individual person, to a human being, in which case you have a completely different understanding of what the agent is, how that agent works, the kinds of data that it's drawing on, the kinds of problems that at solving.
- [02:28]
Peter Ciampelli
Financial services provider BNY is one of those companies that's having success with them. Can you tell me more about how they're deploying them?
- [02:35]
Stephen Rosenbush
They have deployed numerous agents. They have well over 100 AI solutions that touch almost every aspect of the bank, if not every aspect. Some of those AI solutions take the form of agents, and in this case, this was the company that really pioneered the idea of the digital employee. And as their CIO Leon Russell told me, digital workers actually do work, in some cases finding say, problems in software code and even proposing solutions or taking action in simple cases. And that number of those digital employees is actually growing.
- [03:17]
Peter Ciampelli
Walmart is also another company that's using these agents. Can you tell me about how they're using this type of AI?
- [03:23]
Stephen Rosenbush
The AI agents at Walmart fall into a number of broad categories. Customer facing employee or partner facing few others. In one instance, the agent may vet codes, very common application. But they've taken it into some other really interesting territory. The company has used agents that help spot trends. Say what are teenagers doing? What are they interested in, what are they consuming? Say when it comes to apparel. And they'll use the agent at that point to come up with a concept for a product, come up with a design for a product and then serve as an assistant or tool in helping bring that product to market. And they said that the normal process from developing the idea to putting the product on the store shelf or online industry wide takes typically six months. They're able to reduce their time to market by as much as 18 weeks. And he made the point to me that when you have multiple agents working together on a problem like that in that way, that yes, there's a very real return on that investment. No one is saying this is easy, by the way. Leanne Russell at BNY made the point that developing agents wasn't just hard, it was very, very hard. So it's not a simple problem. There's a lot of trial and error. There's, I imagine, plenty of room for failure and it costs money and it requires a lot of expertise. What you're seeing is a proof of concept really at a very large scale in which you're using hundreds of agents distributed among organizations with tens and tens of thousands of employees and they're showing that it's possible to get a return.
- [05:19]
Julie Chang
That was Stephen Rosenbush, the Journal's chief of the Enterprise Tech Bureau, speaking with our Peter Ciampelli. Have you used an AI agent before? If you're a listener on Spotify, be sure to let us know in this episode's poll or leave us a comment. Coming up, Russia was once a tech powerhouse that led the space race. Now it's rushing to catch up in the AI game. That's after the break.
- [05:50]
AWS Narrator
Millions of players, one world, no lag. How's it done? AWS is how.
- [06:01]
Epic games turned to AWS to scale to more than 100 million Fortnite players worldwide so they can stay locked in with battle tested reliability. AWS is how leading businesses power next level innovation.
- [06:23]
Julie Chang
President Vladimir Putin has often proclaimed that Russia must lead the world in artificial intelligence. In reality, the country is stuck on the sidelines as others pull ahead. According to one ranking recently released by Stanford University, Russia languishes at 28th out of 36 countries in AI ecosystem strength. And its leading AI model barely cracks the top 25 worldwide. Here to tell us more about how Russia's AI ambitions stalled is Wall Street Journal foreign correspondent Georgi Khonchev. Georgi, so why is Russia so behind?
- [06:59]
Georgi Khonchev
It's really the Ukraine war that has set back the Russian kind of AI ambitions. Ultimately the Western sanctions that followed the full scale invasion in 2022 have just choked off Russia's access to critical hardware. Computer chips are the most crucial thing here, of course, especially GPUs. Those are difficult to come by in Russia. Russia is impossible to import legally, Nvidia latest models and others. So they have to look for third countries and other suppliers and other kind of hacks to get those models. But that's of course difficult, especially at scale. So Russia ultimately doesn't have access to nearly enough GPUs at the same time, after the beginning of the war, you had the serious brain drain. So a lot of top AI talent tech people left Russia, especially men, to flee the draft. And at the same time, being cut off from international markets means that the whole pool for venture funding in Russia is tiny as well. So overall the main ingredients for AI boom are just non existent in Russia right now.
- [08:06]
Julie Chang
Yeah, that makes sense. So what is the country now doing to catch up to the US or China?
- [08:11]
Georgi Khonchev
They're doing several things. First of all, they're actually depending on China a lot and they're growing even more dependent with time. That's the main thing they can do. And they're getting hardware from China, also some funding, investment and ultimately also they're growing using some of the Chinese AI models as a base. And also Russian companies are trying to again, you know, use kind of ways to get some versions of the latest AI chips by using middleman in third countries. But that's just not nearly enough to get enough in the country. Even stuff like getting a simple subscription to ChatGPT is difficult because Visa, MasterCard and kind of Western payment methods are banned. So what Russian companies are doing, they have to get a provider in Armenia to buy it for them, for example, and then they can use it with a cow to Armenia.
- [09:01]
Julie Chang
What would you say is Russia's biggest challenge in trying to catch up in AI?
- [09:05]
Georgi Khonchev
Ultimately the big issue for Russia is that. Let's go back a bit. The Soviet Union had pretty serious scientific achievements. Obviously, you know about the space race, and they had the computing technology as well. Then Russia, obviously, after the end of the Soviet Union, there was a big economic crisis, but with time, they did manage to develop quite thriving TEC ecosystem. And, you know, in the 2000 and tens, AI was a big topic in Russia. President Putin himself was a big cheerleader of that. So there was a lot of potential in the country. There were a lot of people, a lot of talent, a lot of companies, a lot of startups. There was money going around. And again, on the research and on the development side, there was a lot of promise. And that has evaporated for all these reasons. But talent is especially a big one because you're looking at the top researchers in AI, many of them, according to some estimates, 70, 80% and have left. And then you have IT workers, specialists. It's difficult to come up with a number. There is obviously no number of how many people have left and who those people are. But there's estimates that Russia might be missing up to 400,000 people in the next few years. Some say the real number is much more in the IT sector. But definitely we're talking about hundreds of thousands of people who have left in the IT universe. And that's a big factor here.
- [10:20]
Julie Chang
Georgi, what does it mean for Russia if it doesn't catch up in this AI race?
- [10:24]
Georgi Khonchev
It means several things. I mean, ultimately, we live in a time where obviously AI is transforming the economy and business and how everything is run. But it's also important for various countries to have some form of control over their AI infrastructure, the models, the data, to avoid this dependence on other countries in military too. It's very important that the militaries have their own access to AI capabilities, which are used, as we know, nowadays, daily in the Ukraine war itself. And Putin himself recently has raised that problem. He said, you know, it's very critical for Russia to have its own systems, its own kind of. He said, it's a matter of state. So ultimately it's about sovereignty. And that right now is something that Russia is not able to achieve in the AI sphere for sure, and also in other parts of the economy, too. And it is becoming much more dependent on China, which already is a big trend that's been going on in the Russian economy since the beginning of the war. China is a big buyer of oil, of energy for Russia. It's also supplying with all kinds of tech and cars and other goods. But now AI is becoming one of these sectors where Russia is, to some extent. Totally dependent on outside models, hardware and just outside help.
- [11:32]
Julie Chang
That was WSJ foreign correspondent Georgi Khonchev. And that's it for Tech News Briefing. Today's show was produced by me, Julie Chang with supervising producer Katie Ferguson. We'll be back later this morning with TNB Tech Minute. Thanks for listening.
- [11:53]
AWS Narrator
How do more than 100 million Fortnite players join the battle without lag? AWS is how epic games scales up to keep them in the action. AWS powers next level innovation for millions of businesses.