Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favorite stories in this weekly newspaper.
There was nothing new in artificial intelligence in 2024 that matched the “wow” of using ChatGPT for the first time, but rapid improvements in the underlying technologies kept the field moving. For 2025, that’s how I see things breaking down.
Will AI development hit a wall?
By 2025, that streak will end. Even some stalwarts in the tech industry have acknowledged in recent weeks that throwing massive amounts of data and computing power into training large AI models — a reliable source of improvement in the past — is no longer the case. it begins to bear fruit. diminishing returns. Over time, this deprives AI of a reliable source of improvement. However, at least in the next 12 months, some progress should outweigh the decline.
The most promising developments seem to come from models that run a series of steps before returning an answer, allowing them to question and refine their initial answers to provide “reasonable” results. It remains to be seen whether this can really be compared to human imagination, but systems like OpenAI’s o3 still look like an interesting development since the emergence of AI chatbots.
Google, that is has regained its AI mojo At the end of the year after spending two years struggling to find OpenAI, it also showed how new agent-like capabilities in AI can make life easier, such as tracking what you do in your browser and then offer to complete tasks for you. All these demos and prototypes still need to be turned into useful products, but at least they show that there is more than enough in the lab to keep the AI hype going.
Will an AI ‘killer app’ emerge?
For many people, productivity rises AI It means constantly seeing prompts that help you finish your article or edit your photos in a way you never thought possible – unwanted, useful tools that sometimes just need to change your life. .
Next year is likely to bring the first demonstrations of applications that can directly intervene: receiving all your digital information and learning from your actions so that they can act as memory banks or take over all aspects of your life. But, with concerns about the unreliability of the technology, technology companies will be wary of rushing to adopt it for mass use – and many users will be wary of trusting them.
Instead of real killer applications for AI, this means we’ll be left with the world of “AI in everything” that tech users are already used to: sometimes it’s confusing , sometimes it helps, but it still doesn’t provide new experiences that will prove. The era of AI has indeed arrived.
Are Nvidia GPUs still ruling the tech world?
The chipmaker’s huge profits have made it a target for the most powerful tech companies, many of which are now making their own AI chips. But Nvidia is moving too fast for its competitors, and while a quarter or two may be tough as it goes through a major product transition, its Blackwell product cycles should navigate the coming year with ease.
That doesn’t mean others can’t get involved. According to chipmaker Broadcom, the three largest technology companies should use their in-house designs for the largest “clusters” with 1mn chips each in 2027. That’s 10 times the size of Elon Musk’s Colossus system, which is thought to be the largest cluster. of AI chips currently in use.
Although its market share is beginning to decline, however, Nvidia’s software is still waiting for its big business, and by the end of the year it should be on the verge of another important product cycle.
Will AI’s success in the stock market continue?
With Big Tech in the race for AI that its leaders believe will define the future of their industry, one of the main drivers of AI capitalization will be always there. Also, as some companies start looking for big – if unproven – results from using technology in their businesses, many others will feel compelled to spend money, even if they haven’t worked out exactly how How can I use AI effectively?
Whether this is enough for investors to keep throwing their money into AI is another matter. That will depend on other factors, such as the stock market’s confidence in the new Trump administration’s deregulation and tax-cutting intentions and the Federal Reserve’s willingness to continue monetary policy. .
All point to a very volatile year, with major corrections on the way. But with enough money, Wall Street can give in to the AI hype for a while.
