Friday, March 21, 2025

Welcome to TC PM! Today we look into some tech company experiments, including one Google already performed and one Microsoft is looking to launch. We have more intel on Claude's new web search capabilities, and we learn that apparently people are letting humanoid robots into their house. Couldn't be me. Have a great weekend!

TechCrunch Top 3

Image Credits: Alex Wong / Getty Images

🗞️ Google says no to news: Google recently ran an experiment to see how news affects its ads business. After removing ads for 1% of users in eight markets for two and half months, the company reported that it discovered that news is essentially worthless to its ads business.

📝 They just won't stop: Meta continues to try to push AI onto its social platforms in any way it can. An Instagram user spotted a new feature called "Write with Meta AI," which gives users AI-generated suggestions for things they could comment on posts. Who wants this?

🤖 Humanoids in the home: 1X, a Norwegian robotics startup, said it's going to start testing its humanoid robot, Neo Gamma, in a "few hundred to few thousand" homes by the end of the year. These in-home tests will bring valuable data back to the company.

A Message from Anthropic

Free Anthropic event: How AI is shaping the future for startups

Ready to move your AI projects from evaluation to impact? Join this practical discussion for startups and tech-forward companies that are ready to build and scale AI solutions that deliver measurable business results.

Afternoon must-reads

Image Credits: Gabby Jones/Bloomberg

🎙️ Some words on Wiz: On today's episode of Equity, the hosts dig into Google's recent acquisition of Wiz. They also talk about what Nvidia's GTC conference was like on the ground and dive into Klarna's upcoming IPO.

🤔 Overdue credit: Microsoft is apparently launching a research project with the goal of estimating the influence of specific training material on the media that generative AI creates. This comes as Microsoft is facing two legal challenges regarding AI training on copyrighted material.

💪 Anthropic is brave: Earlier this week we learned that Anthropic's Claude AI chatbot can now search the web. Now we have an idea of what search engine it's using. A software engineer found evidence that Anthropic is using a search engine called Brave Search.

😏 Secrets for scale: Autonomous vehicle startup Wayve's CEO, Alex Kendall, sees promise in bringing the company's tech to market as long as others keep the company's software cheap, tech agnostic, and applicable to different end tech, from robotaxis to robots.

Around the web

🏝️ Trouble in paradise: Everything may not be going as well as it seems for Bryan Johnson, the serial entrepreneur who is trying to live forever. Johnson apparently runs his companies and relationships through a web of confidentiality agreements to prevent gaining a negative image.

💼 This again: The discourse around Section 230, the bill known for preventing internet companies from being liable for what their users post, is swelling up again. A bipartisan pair of senators plans to introduce a bill as early as next week for it to expire in 2027.