AI Agenda by Stephanie Palazzolo — The Information

Four times/week (subscribers only)

AI Agenda

By Stephanie Palazzolo

Stay ahead of the technological breakthroughs, startups and leaders in AI with a 14-day free trial of the AI Agenda. After your trial, unlock 25% off on a subscription to The Information for full access to the newsletter, plus all of our original reporting on AI.

By providing your email, you agree to The Information’s Privacy Policy

Recent Editions

OpenAI Chief Technology Officer Mira Murati introducing GPT-4o. Photo via YouTube/OpenAI.

OpenAI’s Impressive Engineering Feat with GPT-4o AKA ‘Her’

By Stephanie Palazzolo and Kalley Huang · May 14, 2024 7:00am PDT

OpenAI’s Google search killer may not yet be ready for prime time, and neither is the long-awaited GPT-5 large language model. But the ChatGPT developer on Monday may have still managed to steal some thunder ahead of Google’s annual conference for software developers, I/O, which starts later this morning.In a 26-minute demonstration, OpenAI COO... OpenAI’s Google search killer may not yet be ready for prime time, and neither is the...

Sam Altman and Sundar Pichai. Photos via Getty Images.

OpenAI Looks To Steal Google’s Thunder; Unicorn AI Marketing Startup Typeface Acquires Consumer Chatbot App

By Stephanie Palazzolo · May 13, 2024 7:00am PDT

Later today, OpenAI is expected to unveil a bunch of new products. For a sneak peek at what could come, you can check out our story from Friday. But the timing of the event is interesting, given that tomorrow is Google’s largest annual developer conference, Google I/O, where that company is also expected to make a slew of AI-related... Later today, OpenAI is expected to unveil a bunch of new products. For a sneak peek at what could...

Art by Shane Burke.

Developers Push to Build AI Models Native to Non-English Languages

By Kalley Huang · May 9, 2024 8:06am PDT

¿Habla español? Parlez-vous français? You might, but today’s large language models may not speak either Spanish or French very well. It’s a shortcoming that many developers are trying to address, especially since making LLMs available in more languages may attract more users. To give a sense of the issue, Meta Platforms said in a blog post... ¿Habla español? Parlez-vous français? You might, but today’s large language models may not speak...

DatologyAI cofounders, from left to right, Matthew Leavitt, Bogdan Gaza and Ari Morcos. Photo courtesy of DatologyAI.

Ex-DeepMind and Meta AI Researcher Raises Nearly $50 Million To Help Developers With LLMs’ Secret Sauce

By Stephanie Palazzolo · May 8, 2024 7:00am PDT

DatologyAI cofounders, from left to right, Matthew Leavitt, Bogdan Gaza and Ari Morcos. Photo courtesy of DatologyAI.

Ex-DeepMind and Meta AI Researcher Raises Nearly $50 Million To Help Developers With LLMs’ Secret Sauce

By Stephanie Palazzolo · May 8, 2024 7:00am PDT

For all that’s leaked about the sizes and architectures of the advanced large language models developed by OpenAI and its ilk, we still don’t know what data they were trained on.A pessimist might say that’s because of the numerous lawsuits LLM developers would face if the amount of copyrighted content they were using was revealed, but a more... For all that’s leaked about the sizes and architectures of the advanced large language models...

The Rabbit r1 and Humane Ai pin.

AI Devices Get A Rude Wake-Up Call; A Mysterious OpenAI Product Event Is Likely Postponed

By Stephanie Palazzolo · May 7, 2024 7:00am PDT

It’s been hard to miss the absolute beating AI-powered devices have gotten in recent weeks.First, we saw YouTuber Marques Brownlee’s scorching review of the $699 Humane Ai Pin, which he deemed the “worst product” he’s ever reviewed. (Yikes.)Next, the rabbit r1, a kitschy, bright-orange handheld device that first went viral when it launched at... It’s been hard to miss the absolute beating AI-powered devices have gotten in recent weeks.First,...

Photo by Krisztian Bocsi/Bloomberg via Getty Images.

Meet MAI-1: Microsoft Readies New AI Model to Compete With Google, OpenAI

By Aaron Holmes · May 6, 2024 7:20am PDT

For the first time since it invested more than $10 billion into OpenAI in exchange for the rights to reuse the startup’s AI models, Microsoft is training a new, in-house AI model large enough to compete with state-of-the-art models from Google, Anthropic and OpenAI itself.The new model, internally referred to as MAI-1, is being overseen by ... For the first time since it invested more than $10 billion into OpenAI in exchange for the rights...

Photo via Adobe Stock.

How Developers Gave Llama 3 More Memory

By Kalley Huang · May 2, 2024 7:00am PDT

Developers have been praising Meta Platforms’ Llama 3, the latest version of its flagship large language model. But, as my colleague Stephanie and I explained, they have one big criticism: Llama 3’s context window is too short, at just over 8,000 tokens. (As a refresher, a context window is how much information a model can accept in a single... Developers have been praising Meta Platforms’ Llama 3, the latest version of its flagship large...

Photo via Adobe Stock/Shane Burke

Are Publishers Fighting For ‘The Right To Be Forgotten?’

By Stephanie Palazzolo · May 1, 2024 7:10am PDT · 1 comment

Photo via Adobe Stock/Shane Burke

Are Publishers Fighting For ‘The Right To Be Forgotten?’

By Stephanie Palazzolo · May 1, 2024 7:10am PDT · 1 comment

When news broke on Tuesday of a new copyright infringement lawsuit filed against OpenAI and Microsoft, this time by eight daily newspapers, I was reminded of a conversation I had a month ago with Rich Skrenta, executive director at the Common Crawl, a nonprofit that supplies open-source web crawl data.He argued that in the future, consumers... When news broke on Tuesday of a new copyright infringement lawsuit filed against OpenAI and ...

Art by Shane Burke

Are We Running Out of Training Data?

By Stephanie Palazzolo · Apr 30, 2024 7:00am PDT · 1 comment

I was catching up with the new AI Index Report from Stanford’s Human-Centered Artificial Intelligence Institute when one section caught my eye. That was a forecast that we’ll likely run out of high-quality language data needed for train AI models sometime this year.That’s a worrying timeline. So far, AI companies have improved their large... I was catching up with the new AI Index Report from Stanford’s Human-Centered Artificial...

Mistral CEO Arthur Mensch. Photo by Toby Melville via Getty Images.

How Will Open-Source AI Make Money?

By Stephanie Palazzolo · Apr 29, 2024 7:10am PDT · 1 comment

Last week, we spent a lot of time talking about Meta Platforms’ new open-source Llama 3 models. However, the one question that still remains unanswered is how Meta and other developers of open-source AI software will make money from it—an uncertainty that likely contributed to Meta’s stock drop on Thursday.Take Paris-based startup Mistral, for... Last week, we spent a lot of time talking about Meta Platforms’ new open-source Llama 3 models....

Photo by Ronny Hartman/AFP via Getty Images.

Apple’s Late Arrival; Meta Outlines Ways to Make Money from AI, Including Charging Users

By Kalley Huang and Stephanie Palazzolo · Apr 25, 2024 7:10am PDT · 1 comment

If you weren’t careful, you may have missed some important news from Apple that was overshadowed by Meta Platforms’ earnings yesterday. The iPhone-maker released a new family of open-source language models, dubbed OpenELM, on Hugging Face. The models, ranging in size from 270 million to 3 billion parameters (as a reminder, “parameters’... If you weren’t careful, you may have missed some important news from Apple that was overshadowed...

Mark Zuckerberg at Meta Connect 2023. Photo by Bloomberg via Getty Images.

Meta’s Llama 3 Proves a Hit With Developers; Snowflake Gets In The LLM Game

By Stephanie Palazzolo · Apr 24, 2024 7:00am PDT

Mark Zuckerberg at Meta Connect 2023. Photo by Bloomberg via Getty Images.

Meta’s Llama 3 Proves a Hit With Developers; Snowflake Gets In The LLM Game

By Stephanie Palazzolo · Apr 24, 2024 7:00am PDT

It’s been five days since the release of Meta Platforms’ highly-anticipated Llama 3 models, and the new large language models have seemingly won the mindshare of AI developers—so much so that barely anyone was talking about new model releases from Microsoft, Adobe and Amazon on Monday.A number of developers told me that they’re already... It’s been five days since the release of Meta Platforms’ highly-anticipated Llama 3 models, and...

Chart by Shane Burke

AI Investors Turn Their Attention—And Deep Pockets—To Robotics

By Stephanie Palazzolo · Apr 23, 2024 8:15am PDT · 1 comment

If you’re anything like me, you may have found your jaw dropping at our scoop this morning that Skild, a young startup building an AI-powered brain for robots, is raising nearly $300 million from Lightspeed Venture Partners, Coatue Management and other investors at a whopping $1.5 billion valuation despite the fact that it has yet to generate... If you’re anything like me, you may have found your jaw dropping at our scoop this morning that ...

Mark Zuckerberg speaking at Meta Connect 2023. Photo by David Paul Morris/Bloomberg via Getty Images.

Meta Platforms Retakes Its Open-Source Lead

By Kalley Huang and Stephanie Palazzolo · Apr 22, 2024 7:00am PDT

Meta Platforms made a splash Thursday with the release of Llama 3, the latest version of its popular open-source large language model, though we stole some of the thunder with our scoop earlier this month. Developers have a lot to say about it—most of it good.For the most part, the 8-billion and 70-billion parameter versions of Llama 3... Meta Platforms made a splash Thursday with the release of Llama 3, the latest version of its...

Art by Clark Miller.

What To Be Excited About For AI in 2024; How Much Power Will Stargate Need?

By Stephanie Palazzolo · Apr 18, 2024 7:30am PDT

When you’re covering a space as hyped-up as generative AI, it can be easy to become jaded and overly skeptical (something I have to combat every time I see yet another AI funding round valued at a 100-times forward revenue multiple).But, there’s still plenty in AI to be excited about—just take a look at this piece from my colleague Aaron Holmes,... When you’re covering a space as hyped-up as generative AI, it can be easy to become jaded and...

Photo via Midjourney.

Why Many AI Startups Are Consultancies Posing as Software Businesses

By Stephanie Palazzolo · Apr 17, 2024 7:00am PDT · 3 comments

Photo via Midjourney.

Why Many AI Startups Are Consultancies Posing as Software Businesses

By Stephanie Palazzolo · Apr 17, 2024 7:00am PDT · 3 comments

There are plenty of reasons generative artificial intelligence hasn’t taken off with businesses as much as pundits expected, including high costs and errors (i.e. hallucinations). One less-discussed reason is that conversational AI is simply hard to use. Big companies outside of tech can’t often hire the kind of technical talent needed to make... There are plenty of reasons generative artificial intelligence hasn’t taken off with businesses...

Pay no mind to the man behind the curtain. Photo via YouTube/Warner Bros. Entertainment.

When Does ‘Fake It Till You Make It’ Just Become ‘Faking It’?; The History of ‘Plagiarism’ in AI

By Stephanie Palazzolo · Apr 16, 2024 7:00am PDT

Fame can be a double-edged sword—a lesson some artificial intelligence startups are learning the hard way.Take Devin, the AI-powered coding assistant developed by Cognition AI that went viral when it launched last month with an impressive demo showing how it could complete coding jobs on the freelance site Upwork. Techies across X and Reddit... Fame can be a double-edged sword—a lesson some artificial intelligence startups are learning the...

Thomas Kurian, CEO of Google Cloud. Photo via Google.

Google Cloud’s AI Strategy Is Starting to Sound Like AWS’

By Anissa Gardizy · Apr 15, 2024 7:00am PDT

I’m back in San Francisco after a busy week in Las Vegas, where my colleague Jon Victor and I spent days talking to Google Cloud customers about the difference between generative artificial intelligence hype and reality. At one point during the conference, I was sitting at a table in a chilly Mandalay Bay conference room when I heard a... I’m back in San Francisco after a busy week in Las Vegas, where my colleague Jon Victor and I...

Photo via Getty Images.

A New Wave of Foundation Model Developers Is Coming

By Stephanie Palazzolo · Apr 11, 2024 7:20am PDT

For months, venture capitalists have been telling me the same thing: the foundation model game is over. Between OpenAI, Anthropic, Cohere, Google and dozens of open-source model developers, there’s just no way for a company developing a new large language model to break into the market. In recent weeks, though, VCs have modified that stance... For months, venture capitalists have been telling me the same thing: the foundation model game is...

Cris Valenzuela, CEO of Runway. Photo by Nicole Pereira.

Don’t Mortgage Your Future to Microsoft, Says Coatue’s Chief Investment Officer—And What Else You Missed from Our Private Capital Conference

By Stephanie Palazzolo · Apr 10, 2024 7:00am PDT

Cris Valenzuela, CEO of Runway. Photo by Nicole Pereira.

Don’t Mortgage Your Future to Microsoft, Says Coatue’s Chief Investment Officer—And What Else You Missed from Our Private Capital Conference

By Stephanie Palazzolo · Apr 10, 2024 7:00am PDT

Artificial intelligence wasn’t the stated theme of The Information’s Private Capital Conference on Tuesday at the New York Stock Exchange but it was the topic on everyone’s lips. Artificial intelligence wasn’t the stated theme of The Information’s Private Capital Conference...