|
Thanks for reading The Briefing, our nightly column where we break down the day’s news. If you like what you see, I encourage you to subscribe to our reporting here.
Greetings! If anyone enjoys this holiday, it should be Jensen Huang. Nvidia stock on Thursday closed at $159.34 a share, up 18.6% for the year, giving it a market capitalization of $3.88 trillion. Getting to $4 trillion is within reach. (Remember when passing $1 trillion was a big deal for any company?) Right now, Nvidia is valued only a little below what Alphabet and Meta Platforms are worth on a combined basis. Let that sink in. There’s no puzzle what’s going on. After a stock wobble prompted by President Trump’s tariffs, specifically his order blocking Nvidia from selling artificial intelligence chips to China, investors have flocked back to the AI chip giant. A Morgan Stanley report on Monday, citing “very strong demand for all Blackwell form factors”—Nvidia’s latest AI chip—helps explain the enthusiasm. Moreover, Nvidia last month announced a variety of agreements with European AI model developers, suggesting the European market was expanding, perhaps helping offset a little what it's losing in China. Nvidia faces lots of would-be rivals, including from many of its customers, but so far there’s little sign anyone is a serious threat, at least right now. Microsoft, as we reported on Wednesday, has scaled back its ambitions to develop its own AI chip after having to delay those it was working on. Google uses its internally designed tensor processing units for training and operating most of its AI models but still buys a lot of chips from Nvidia for customers of Google Cloud. And while Google managed to get OpenAI to test its TPUs, The Information’s report last Friday about OpenAI’s use of the chip seemed to have caused ripples throughout the industry. Two days after the story ran, OpenAI suddenly put out a statement to Reuters saying it had been testing TPUs but that it had “no plans to deploy them at scale right now.” You have to wonder what went on behind the scenes there. Meanwhile, as we reported today, Apple has succeeded in using its own chips for certain AI capabilities, but it’s far from a threat to Nvidia. The long term demand trajectory for Nvidia’s chips remains a question mark. But even at its current stock price, Nvidia shares don’t appear overly pricey. The stock trades below Broadcom on a forward sales basis, and below Broadcom and AMD on a forward multiple of earnings, according to S&P Global Market Intelligence. The company will pass $4 trillion in market cap once it hits $164 a share. Is that next week? OpenAI has been losing researchers to Meta—last Saturday we scooped the latest Meta hires—but the ChatGPT creator remains in the premier position in AI development. That extends to the sale of AI services to business. Last weekend we dug into how the firm is pitching its AI services to big companies and government agencies—by offering in-house researchers and software engineers to customize AI for them. Also on the AI front, we revealed that Anthropic has grown so fast that it has reached an annual revenue rate of $4 billion, up almost four times from the start of the year. Our AI Agenda columnist Stephanie Palazzolo questioned why Meta was bothering to pursue superintelligence when a lower grade of AI would suit its purposes just as much. We broke the big news that Apple had explored launching a cloud service for developers in its iOS and Mac ecosystems, offering to rent servers powered by Apple chips. Competition is intensifying between enterprise software firms and AI firms over access to business customers’ data. This story is worth reading to understand the backdrop between the Salesforce-Glean drama of a couple of weeks ago. Outside of the AI world, we looked at the intense competition for a share of the $19 billion streaming TV ad market in the U.S. One group of competitors, Free Ad Supported Television channels (FAST for short) has been squeezed by premium players like Amazon, we reported. Vice President JD Vance is meant to be one of President Trump’s key tech advisors. But his views on tech can be hard to pin down, as Sylvia Varnham O’Regan reported from Washington, D.C. • Buyout firm Thoma Bravo has agreed to acquire restaurant software maker Olo for $2 billion, the companies announced on Thursday. • Safe Superintelligence, the secretive AI lab co-founded by OpenAI co-founder and former chief scientist Ilya Sutskever, confirmed the departure of co-founder and venture capital investor Daniel Gross. Sutskever is now CEO and will continue to oversee the company’s technical team. Gross is in advanced talks to join Meta’s AI efforts. The Information Weekend covers what happens when Silicon Valley logs off—the trends and people shaping culture, technology and everything in between. Subscribe for free today. | 
