Daniel Howley
It’s been a little more than a year since OpenAI’s ChatGPT hit the web, setting off an explosion in interest in generative AI. In the months since, tech giants including Microsoft (MSFT), Google (GOOG, GOOGL), Amazon (AMZN), Meta (META), and others have debuted or announced they’re working on their own generative AI chatbots and products.
But that’s so 2023.
We’re interested in what we can expect from generative AI in the year ahead. What kind of surprises does 2024 have in store for the tech industry — and how will generative AI influence them?
“2024 is going to be the year when it really explodes, because every day people are going to use [AI],” TECHnalysis president Bob O’Donnell told Yahoo Finance.
Think PCs and smartphones running generative AI programs, and generative AI-powered video and audio platforms.
That’s not all, though. According to experts, generative AI will become more targeted. Systems like ChatGPT, Google’s Bard, and Microsoft’s Copilot — previously called Bing Chat — are general purpose platforms. They’re basically designed to answer everyone’s questions. But they’re not experts in individual fields.
OpenAI CEO Sam Altman, left, appears onstage with Microsoft CEO Satya Nadella at OpenAI's first developer conference, on Nov. 6, 2023, in San Francisco. (Barbara Ortutay/AP Photo, File) (ASSOCIATED PRESS)
In the coming year, companies will begin releasing models that are tailored to specific tasks, rather than those meant for broad uses.
The idea is to make generative AI less of a jack of all trades and more a master of one. Add to that the continued proliferation of more generalized AI models, and 2024 is shaping up to be another year dominated by AI.
Generative AI for consumers
While generative AI was the talk of 2023, there are still huge swaths of the world who haven’t used the technology. But as generative AI continues to permeate more of the services and products we use, you can expect more people to gain access to those platforms, said Daniela Rus, director of MIT’s Computer Science and Artificial Intelligence Laboratory.
“With the major foundational models that are working increasingly better, we really do see a democratization of AI and machine learning,” Rus told Yahoo Finance.
A series of servers powering Google's Gemini AI platform. (Google) (Google)
“We really see that a lot of people with access to a phone or a computer are now able to take advantage of what the technologies can do for them. And so, I see increasingly more people using the tools for their own personal purposes,” she added.
One way in which people will gain access to generative AI capabilities will be through new consumer hardware outfitted with AI chips.
“2024 is going to see the launch of AI-equipped PCs; PCs that actually can do some of the work that we’ve traditionally had to do in the cloud, on device, and that’s going to make things very interesting,” O’Donnell explained.
Google is already making such moves. In October, the company launched its Pixel 8 and Pixel 8 Pro smartphones. The devices, which run on Google’s custom Tensor G3 chip, feature powerful AI capabilities including the ability to alter photos, remove background audio from videos, and generate wallpapers.
In the year ahead, companies ranging from HP and Dell to Lenovo and perhaps even Apple should begin integrating new AI functionalities into their devices.
AI gets more specialized
AI platforms like ChatGPT are designed to be general purpose systems. Throw a question at it about the 1997 New York Jets, and it’ll spit out an answer. Want to write a poem about leftover lasagna? It’ll do that too.
Meta CEO Mark Zuckerberg delivers a speech at the Meta Connect event at the company's headquarters in Menlo Park, Calif., Sept. 27, 2023. (Carlos Barria/REUTERS) (Carlos Barria / reuters)
But 2024 will bring more specialized generative AI platforms aimed at specific topics and projects.
“I actually think we'll see … less general-purpose stuff,” explained University of Maryland computer science professor Hal Daumé III.
“My nose would say go where the data is, because these models are all super data hungry. So … if I had to make predictions, I would start thinking about what are the other places where we have huge amounts of data,” Daumé added.
Types of specialized generative AI applications could include platforms designed to help improve weather predictions, cybersecurity services, and medical research.
Meanwhile, chip competition heats up
Nvidia (NVDA) is the most important chip company in the world right now. It not only develops the AI accelerators that power some of the biggest AI systems in the world, the firm also offers the software needed to develop them.
But Nvidia doesn’t have enough chips to go around, so tech giants like Microsoft, Amazon, Google, and others are developing and using their own custom AI chips. And while that might not eat into Nvidia’s bottom line, increased competition from rivals AMD (AMD) and Intel (INTC) could.
“We’re going to see a ton of other semiconductor companies … going after Nvidia,” said O’Donnell. “And the truth is, the market is looking for more competition. It always does, right?”
AMD and Intel aren’t the only ones setting their sights on Nvidia, though. Qualcomm (QCOM), which is known for its mobile chips, is also angling to become a larger presence in the AI space.
Simply put: If you thought generative AI was huge in 2023, generative AI in 2024 just might blow you away.
Daniel Howley is the tech editor at Yahoo Finance. He's been covering the tech industry since 2011. You can follow him on Twitter @DanielHowley.
No comments:
Post a Comment