🚀 Altman: More leaps coming

Secretary of State Antony Blinken — a UN General Assembly veteran — passed along what he said is his No. 1 piece of advice for making the most of the event: "Take the subway."

Today's AI+ is 1,225 words, a 4.5-minute read.

To those who contend that large language models are dumb word-predictors — "stochastic parrots" — that don't actually understand questions or solve problems, OpenAI CEO Sam Altman responds, in effect: So what?

The big picture: As the power of generative AI models continues to grow and the tech industry's bets on the technology pile up, a gulf remains between industry leaders like Altman — who believe AI will keep getting better as it matures — and critics who argue that the technology will never prove fully reliable.

What they're saying: "I think people get very hung up on the fact that it's just being trained to predict the next token," Altman said yesterday in an onstage interview with Axios in New York.

• "Once it can start to prove unproven mathematical theorems, do we really still want to debate 'Oh, but it's just predicting the next token?'" Altman said.
• He spoke on a panel with UN tech envoy Amandeep Singh Gill at an OpenAI event on the sidelines of the UN General Assembly meeting this week.
• The interview followed a rosy-eyed blog post that Altman published Monday morning laying out a vision of a new "Intelligence Age," in which AI will bring "shared prosperity to a degree that seems unimaginable today."

Altman suggested that we've quickly taken for granted achievements most of us would have thought impossible just a few years ago.

• The recent release of o1 (formerly code-named Strawberry), he said, lets anyone on the internet use a tool that's better at math than all but the top few 100 students in the U.S., and works at the level of upper-echelon programmers.
• Altman recalled an observation by mathematician Terence Tao: "The thing [he] said that stuck with me is that before, AI was like a very incompetent grad student. Now it's like a mediocre grad student that you could give tasks to. And soon, you can see a unique, useful research partner."

State of play: Altman said the new o1 model's reasoning capability has convinced him AI is getting closer to the goal of advancing scientific discovery.

• "I think with o1, you can see the glimmers of how this is starting to work," Altman said.

The other side: Altman's comments stood in sharp contrast to the perspective provided by the UN's Gill, who talked about challenges he had seen in past work using AI to address global health issues.

• "I don't think we can get a shortcut to all of this through large language models," he said.

Between the lines: Gill also lamented that people are using such compute-intensive systems to do simple tasks that could be better accomplished using other means — like just asking other people for help finding the nearest Starbucks.

Yes, but: Altman argued that such queries represent a trivial use of energy today, will use even less energy in the future and are accompanied by other types of inquiries that save time and energy because they are solved by AI.

• Altman pointed out that costs for a set AI computing task have rapidly come down.
• He added that when OpenAI and its rivals cut prices, that's because they have found ways to accomplish the work using less computing — and therefore less energy, too.

OpenAI CEO Sam Altman confirmed yesterday that the hardware project he is working on with former Apple design chief Jony Ive — rumored to be a new kind of AI-driven device — is not a phone.

What they're saying: "I don't think you should try to do a better phone," Altman told Axios at the same interview.

State of play: Whatever the product is, it won't be coming any time soon.

• "It's a long way away," Altman said, noting that it took OpenAI more than four and a half years to ship its first product, "and I thought that was fast."

Catch up quick: Ive, who has long been reported to be working with Altman on a hardware project, confirmed the collaboration in a recent New York Times interview.

• For his part, Altman noted last year that major platform shifts usually usher in a new type of computing device. He told me then that "if there's something amazing to do, we'll do it."
• Altman previously invested in AI hardware startup Humane.

The big picture: OpenAI has been pushing forward along multiple paths, including the recent release of its o1 reasoning model and advanced voice chat with GPT-4o.

• It's also said to be close to finalizing another round of financing in the vicinity of $6.5 billion, which would be the largest venture capital round in tech history, per Axios' Dan Primack. Altman declined to comment on that.

Altman said OpenAI isn't done with big announcements this year, though he declined to say if the long-awaited GPT-5 is on tap — or even if that will be its name.

• "Expect some more cool things from us this year," Altman said. "Let's say that."

Google is holding a "Gemini at Work" event today to convince businesses that its generative AI is better than offerings from Microsoft and OpenAI.

Why it matters: The largely virtual event comes amid a flurry of claims from tech providers and growing skepticism that genAI is ready for broad use beyond coding and customer support.

Driving the news: Google is using the event to tout more than 100 examples of customers using its Gemini and Vertex AI tools for a wide range of business purposes.

• That list includes Snap, Dun & Bradstreet, Puma and Radisson Hotel Group.
• Scotts Miracle-Gro, for example, is using Google Cloud's Vertex to create an AI "garden sommelier" that dishes out advice for aspiring green thumbs.
• The company is also announcing some modest product enhancements and talking up the latest trend in genAI — semi-autonomous agents that can take action within prescribed limits.
• Salesforce and Microsoft have both highlighted their agent efforts in recent days.

Between the lines: One of the key questions for businesses is whether the productivity assistants offered by Microsoft, Google and others are worth paying an extra monthly fee of $20 to $30.

• Even modest productivity gains could offset the cost.
• Google says its large business customers are saving an average of 105 minutes per person per week.

Zoom in: In a test, the Clearwater, Florida-based moving company Pods outfitted its storage and moving trucks in the New York area with digital displays and used Google's AI to generate slogans that included local facts.

• "People always know us and have known us as storage in your driveway because of that container," Pods VP of brand and media Calvin Fields told Axios. "And so we're like, what if we use that container to be a billboard?"
• Training Gemini took some upfront effort, Fields said, but in the end Pods found the AI could generate the content much faster than people. (Humans did check Gemini's work, Fields said.)

Is AI, like internet search and email, going to consolidate into a handful of dominant players?

"If one of them becomes the clear, dominant leader, then it's going to have incredible economics," says Sung Cho of Goldman Sachs Asset Management.

Read more.

• Microsoft issued a report outlining what it says is progress it has made in security following a series of incidents and reports critical of the company's practices. (Axios)
• A new AI startup using AI to transcribe and summarize pet appointments for veterinarians has raised an $8.2 million seed round led by Andreessen Horowitz. (Axios Pro)

I was feeling judged, or at least seen, as I walked by this cafe in Manhattan yesterday.

A tide of investment is pouring into generative artificial intelligence. Will it be worth it?

To see where the industry is headed, portfolio managers from Goldman Sachs Asset Management met with executives from 20 leading technology companies driving AI innovation.

Get the analysis.

Thanks to Megan Morrone and Scott Rosenberg for editing this newsletter and to Caitlin Wolper for copy editing it.