Technology

➚ MOVE FAST: Big Tex. With U.S. chip production ramping up, Samsung is set to double its investment in the Lone Star state to $44 billion. It will add to the semiconductor hub it’s already building near Austin, with funds from the CHIPS Act expected to help finance the latest project.

➘ BREAK THINGS: Big Apple. New York City’s new AI chatbot to help local businesses is giving out dubious information. It told employers it’s ok to take a cut of workers’ tips and landlords could essentially reject potential tenants based on their income. Mayor Eric Adams defended the technology, and said “kinks” needed to be ironed out.

Marissa Mayer is co-founder of Sunshine, an app development company that recently launched a new photo-sharing service, Shine. She was CEO of Yahoo and a longtime Google executive.

Q: You started off with a contacts app and now you have a photo app. Can you explain the strategy?

A: Photos and contacts go together. It’s obvious when you think about it, but the contacts you’re closest to, you tend to spend time with, and if you spend time with them in this day and age you have photos of and with them. So you can understand people’s relationships if you can look at their camera roll. And now with the beauty of AI, we can actually analyze someone’s camera roll really succinctly and have an understanding of who you spend time with, when, where and why, at least in broad brushstrokes.

That’s really useful when it comes to keeping people in touch with each other, helping take those mundane, everyday touches and make them magical and really helping people feel connected to each other. Contacts are a big part of that. It was the foundation and that’s why we started there. But photos are perhaps one of the most meaningful ways that we all interact with each other. The way we all build those relationships.

Q: It resonates with me because I’ve never loved social media but I do want to share with people I know. Do you see consumer tastes changing these days in the post-web 2.0 era?

A: Yes, maybe. I also think that AI and our expectations from it are going to influence it. One way to think of it is if Instagram is one perfectly filtered, edited photo that you push out to everyone you know and don’t know who is in your follower base, and person-to-person photo sharing is me snapping a picture and texting it or air dropping it to you, Shine tries to sit in that space in between. The ‘okay, I was just somewhere with 10, 20, 50 people. And it was an interesting night, but I didn’t kind of want to take myself out of the moment and take a photo. But I know I took some, they took some, let’s just make a big photo pool of what that trip, party, conference, event felt like.’ That’s really the space we’re trying to occupy.

There’s not a lot happening in that space. I refer to it as smart, small-scale sharing. The early social networks and early social media tried to fill that space. But then people would join, they wouldn’t see a lot of the people there that they wanted to share with. So they looked for who is famous, which is why you suddenly started to get these counts into the millions for celebrity followers.

But at the same time, celebrities are interesting, but so are your friends, families and your close contacts and they’ve got stuff going on in their lives that they want to share with you, too. That’s really the space we want to play in. I think Snapchat showed this in a different way — you can share in a way that’s a little bit less perfect, and a little bit temporary and in the moment. There’s something really powerful about that.

Q: As somebody who started out in symbolic systems, do you geek out on the architecture of these AI models?

A: The thing that I’ve found the most surprising and delicious in all of this is symbolic systems. Symbolic systems at Stanford is a unique major. It’s basically cognitive psychology with a really heavy CS bent. But the actual makeup of the major is cognitive psychology: How do people learn? Philosophy: How do they reason? Linguistics: How do they express themselves? And computer science: Can you create a computer that can learn reason and express itself?

I was fascinated by this in the mid 90s. Both my degrees at Stanford are in the concentration or a specialization in AI for my bachelor’s and my master’s, respectively. I thought that was very interesting and there’s all these symbolic systems. We see that thread, learn, reason and express yourself.

I loved computer science from the very first time I coded. I didn’t know much about linguistics. I understood it was important for the intelligence to express itself, but my focus and I will say overall, the industry’s focus, has over the past decade or two, been much more on learning. Building the models, reinforcement learning, all of the different techniques that go into building a modern day model right now, and then deploying those models, giving them new things to reason about and see how they perform. That’s where all the energy is gone.

And the big surprising thing that happened with OpenAI and ChatGPT, at least for me, was that expression piece, that linguistics piece, that was perhaps under-appreciated in that cycle of symbolic systems is what captured everyone’s imagination, the fact that it can talk and express itself and write in a way that you can understand and feels really intelligent.

The fact that Dall-E can draw these wonderfully complicated pictures that would take an artist days or weeks to express. That really captured the imagination. So it’s funny because for a long time, I didn’t understand and kind of discounted that expression piece. And in this moment, it’s that expression piece that has been the big unlock for AI, and really helping everyone see its potential.

The number of employees Apple reportedly let go recently, likely in relation to its car program fizzling out. To put that into context, Tesla, for instance, employs more than 143,000 people.

Your daily dose of what’s new in emerging tech. MIT Technology Review’s daily newsletter, The Download, helps you navigate the latest technologies, why they matter, and how they impact the world around us. Sign up for free today.

Google is considering charging for some of its AI-powered software, the FT reported. On the one hand, it would be surprising if Google is not, on some level, considering literally everything as its core business comes under threat from technological disruption. On the other hand, the news highlights how the business model of the internet could be upended by AI.

The conundrum for Google is that we’re in an awkward stage right now, where the technology isn’t really that useful. A small handful of consumers may be willing to pay for ChatGPT’s premium models, but the vast majority of people view it as a kind of novelty. Google will not, for a very long time, be able to earn enough money via consumer subscriptions to have a meaningful impact on its top line.

It’s probably time for Google to rethink its mission of organizing the world’s information. In some ways, Google’s bigger impact was not so much search, but creating the business model that funded the web.

From that perspective, the way Google could win in the generative AI era is by inventing the infrastructure layer that sits between language interfaces and traditional software-enabled services — and then figuring out financial incentive structures that could fund it. In other words: The DoubleClick of the LLM era.

Yesterday was a big day for Cohere, the enterprise AI company that has been operating in the shadows of flashier startups like OpenAI and Anthropic. Cohere released a new, more powerful AI model Thursday, called Command R+. Perhaps more significantly, though, it announced it was now available on Microsoft Azure. This is kind of like Target agreeing to stock your products on its shelves. It doesn’t guarantee sales, but it certainly boosts your chances.

This is relevant because Cohere’s revenue growth was called “tepid” by The Information recently. In an interview with Semafor on Thursday, Cohere president Martin Kon said the company’s revenue has grown 60% this year. “And now being on Azure, the momentum is just going to be absolutely huge,” he said.

Here’s why I think 2024 presents a good opportunity for Cohere: They aren’t trying to do anything too fancy. As Kon loves to point out, they don’t have a consumer chatbot. They’re focused on retrieval augmented generation, or RAG. That means using a large language model to search only certain data for its answers — usually on a company’s servers. The idea is that it reduces the chances for hallucination.

Right now, AI models just aren’t that good. Even OpenAI CEO Sam Altman said last week that the company’s most advanced model “kind of sucks.”

Until AI doesn’t suck, the scope of its work needs to be limited and narrowed. I asked Kon if Command R+ with RAG is essentially a hallucination-free service. He didn’t definitely answer the question, but he points out that when Command R+ responds, it cites its source, so customers can verify the data.

I haven’t used Cohere’s product but the messaging makes sense. AI companies need to promise less and deliver more.

• The border crisis might be a boon for the economy.
• Unions’ picket power now extends to U.S. boardrooms.
• Ethiopia bets on property ownership offer to attract foreign investors.