To get a roundup of TechCrunch’s biggest and most important stories delivered to your inbox every day at 3 p.m. PST, subscribe here.
Hello, Crunchers! We’re pretty excited about Matt’s TechCrunch Live, where he talks to Cambly about how the company found profits after failing to raise a Series A. Mark your calendar for February 1! — Christine and Haje
The TechCrunch Top 3
- And we’re back!: That’s what Microsoft is saying after some of its services, like Outlook, Xbox Live and Teams, went down during a “change made to the Microsoft Wide Area Network,” Ivan reports.
- What you text is what you’ll see: We’ve been hearing about text-to-image lately, and today is no different. Remember in October when Shutterstock and OpenAI paired up to add some artificial intelligence to Shutterstock’s libraries? Well, today the stock photo giant is showing us the fruits of that labor with a generative AI toolkit to create images based on text prompts. Ingrid has more.
- Plugged in: BMW i Ventures gets a renewed charge from one of its newest investments, infusing $13 million into Bulgaria-based Ampeco, a company providing an electric vehicle charging management platform, Mike reports. As he notes, you might recall that BMW was an early investor in exited companies ChargePoint and ChargeMaster.
Startups and VC
Injective, a layer-1 blockchain focused on building financial applications, has launched a $150 million fund ecosystem initiative, the platform’s CEO and co-founder, Eric Chen, told Jacquelyn in her article, Injective launches $150M ecosystem fund to accelerate interoperable infra and DeFi adoption.
One of the most remarkable things about construction robotics is the sheer breadth of tasks that can potentially be automated, Brian writes. He believes the entire category is a prime target for robotics startups, given that it addresses the three big Ds of automation — dull, dirty and (quite often) dangerous. It makes sense, then, that Built buys fellow construction robotics firm, Roin.
Fun stuff. There’s five more, too:
When it comes to large language models, should you build or buy?
Image Credits: Jenny Dettrick (opens in a new window) / Getty Images
Americans spent nearly $20 billion on pizza deliveries in 2021. Most people could probably bake one at home, but speed and convenience are powerful incentives at dinnertime.
The same holds true for machine learning algorithms: Should companies select open source models, license large language models without modifications, or customize them and pay much higher usage rates?
“While building looks extremely attractive in the long run, it requires leadership with a strong appetite for risk over an extended time period,” writes ML engineer Tanmay Chopra.
Three more from the TC+ team:
TechCrunch+ is our membership program that helps founders and startup teams get ahead of the pack. You can sign up here. Use code “DC” for a 15% discount on an annual subscription!
Big Tech Inc.
Frederic took a look at what Google was doing at this year’s Flutter Forward event and found that the open source framework got some new graphics capabilities, and is launching its first efforts to compile Flutter to WebAssembly and is working on some RISC-V support. He writes, “Virtually all of these capabilities still sit in canary branches and behind experiment flags, but they do show where Google plans to take this project in the months ahead.”
Now here’s five more for you: