The UK is looking into Microsoft and OpenAI’s partnership

Hello, fellow humans! You’re reading our limited series of Saturday Daily Briefs. While it’s focused on AI, it’s curated, written, and edited by actual people.

Got some questions about AI you’d like answered? Or just some AI hallucinations you’d like to share? Email us anytime. Enjoy!

Correction: Friday’s Daily Brief incorrectly referred to Viktor Orbán as the prime minister of Turkey. He is the prime minister of Hungary.

Here’s what you need to know

Intel has a new AI chip. The company’s Gaudi3 will come out next year to compete with rival tech made by Nvidia and AMD.

OpenAI’s nonprofit arm made just $45,000 last year. With private investors valuing the company at $86 billion, its financials remain a black box.

Meanwhile, one of OpenAI’s competitors in Paris raised €385 million ($415 million) in its latest funding round. Mistral AI’s valuation, however, is $2 billion—a far cry from the ChatGPT maker’s.

The parent company of Business Insider and Politico struck a content agreement with OpenAI. The deal will allow ChatGPT to use articles from Axel Springer’s publications to answer user prompts and provide links to those stories.

Antitrust in an AI world

The UK’s antitrust agency, the Competition and Markets Authority, is gathering information on Microsoft and OpenAI to decide whether their partnership threatens competition in the UK, which Google’s AI subsidiary DeepMind calls home.

The fast-moving generative AI industry is increasingly catching the attention of regulators globally. Right now, the CMA is just seeking comment from the firms, but that move could lead to a formal antitrust investigation by the British agency.

Greater antitrust scrutiny is making it harder for AI-focused companies to join forces. Amazon, Google, and Meta have responded by finding ways to work with startups rather than buy them outright, Brendan Burke, an analyst at PitchBook, previously told Quartz. But after OpenAI’s recent turmoil, a spotlight has been put on its ties with Microsoft that’s hard to dim.

Google is doubling down on live-action AI

When Google dropped its new generative AI model Gemini in a prerecorded video demo last week, it highlighted the differentiating elements of the ChatGPT rival, such as its ability to converse out loud. But there seemed to be a gap between what was shown and the large language model’s true capabilities.

Sure enough, Google admitted the video was edited. The YouTube description of the demo said “latency has been reduced and Gemini outputs have been shortened for brevity,” which makes it look like the AI system is responding faster than it is.

The staged aspects of the demo got called out by sites like Bloomberg and TechCrunch, not a good look for the ChatGPT rival that’s racing to catch up with OpenAI. Google must have taken note, returning this week with live demos of its Gemini Pro and other AI tools. The change in approach underscores the need for Google to build confidence in its generative AI capabilities—but will investors take note?

Quotable: Putin, meet Putin

“Vladimir Vladimirovich, hello, I am a student at St. Petersburg state university. I want to ask, is it true you have a lot of doubles?”—An AI-generated version of Russian president Vladimir Putin asked the real Vladimir Putin during an annual news conference on Thursday.

The real Putin hesitated briefly, then said: “I see you may resemble me and speak with my voice. But I have thought about it and decided that only one person must be like me and speak with my voice, and that will be me.”

Other great AI reads

🙏 The Pope, once a victim of AI-generated imagery, is calling for a treaty to regulate it

🫣 2023 was the year we played with AI, and weren’t sure what to do about it

💻 Oracle literally can’t build enough cloud computing for Elon Musk’s xAI

🇪🇺 Europe agreed on world-leading AI rules. How do they work?

🇲🇾 Nvidia’s CEO thinks that Malaysia could be an AI manufacturing hub

☁️ AWS chief Adam Selipsky talks AI, Anthropic, and cloud cost cutting

Ask an AI

Earlier this week, we told you about how Tesla had to recall nearly 2 million vehicles in North America to fix its Autopilot feature. The company is adding more features to ensure drivers are paying attention, but safety experts say it isn’t enough. The real problem, they maintain, aren’t drivers, but the technology itself, which has issues with spotting obstacles and shouldn’t be allowed on the road.

We asked the Grok AI (only available with an X Premium+ subscription), another Elon Musk product, to weigh in on the issue of whether self-driving cars are safe. It said, “I believe that self-driving car technology has come a long way and is generally safe. However, like any technology, it’s not perfect, and there have been instances where accidents have occurred. As a humorous AI, I’d say that self-driving cars are like teenagers learning to drive—they’re pretty good most of the time, but every now and then, they might make a questionable decision. In all seriousness, though, it’s essential for companies to continue improving the technology and addressing safety concerns to ensure a smooth and secure ride for everyone on the road.”

Grok self-identifies as humorous, it pulls its information from tweets, and it’s still in its infancy, so it can’t really be considered the most useful or reliable AI out there. But it’s hard not to read the above quote and raise an eyebrow or two.

When we asked Grok how an autopilot feature would keep a driver and the people around it safe, it did have this helpful tip: “While the autopilot feature can help with steering, it’s essential to maintain control of the vehicle at all times. Think of it like holding hands with a friend—you’re there to support each other, but you’re still in charge of your own actions.” Aw!

Did you know we have two premium weekend emails too? One gives you analysis on the week’s news, and one provides the best reads from Quartz and elsewhere to get your week started right. Become a member or give membership as a gift!

Our best wishes for a very human day. Send any news, comments, driving tips, and Putin calendars to talk@qz.com. Reader support makes Quartz available to all—become a member. Today’s AI in Focus Daily Brief was brought to you by Morgan Haefner and Susan Howson.