Welcome back to review week! There are plenty for you this week, including those that came out of WWDC 2025. AI browser from a browser company. Partnership between Openai and Mattel. iPad update. Have a great weekend!
Apple Experience: We started the week at WWDC 2025, Apple’s Worldwide Developers Conference. There, they showed off the newly designed iOS 26, new features from the entire product, and more. This year, Apple is behind AI and relies on the App Store.
Snack Hack: United Natural Foods (UNFI), the leading US grocery retailer, has been hit by a cyber attack, the company confirmed Tuesday. Many of UNFI’s external systems were offline, including web systems used by suppliers and customers, as well as the company’s VPN products. Whole Foods was one of the victims, and the cyberattacks have affected UNFI’s “ability to select and ship products from warehouses,” which “impacts normal delivery schedules and product availability.”
Public Debut: Chime’s highly anticipated public debut has finally arrived, with the company raising $864 million in IPO. Iconiq was one of many supporters of Chime who won to become a public company upon graduation.
This is TechCrunch review week, summarizing the biggest news of the week. Want to deliver this to your inbox as a newsletter every Saturday? Sign up here.
news

Don’t lose: Google has deployed Android 16 to Pixel phones, added group chat to RCS, suggestions for AI-powered editing in Google Photos, and support for corporate badges in Google Wallet.
The taxi is here: Elon Musk has spent years claiming Teslas can drive himself. Apparently the time has come – maybe? Musk said Tesla will begin offering public vehicles on unmanned vehicles in Austin, Texas on June 22.
AI Browser: The browser company said last year it would be halting support and development of ARC browsers. The startup has since been busy developing an AI first browser called DIA.
Another: Openai has released the O3-Pro. This is the version of O3, an inference model that started by the startup earlier this year. In contrast to traditional AI models, inference models work staged problems, allowing them to run more reliably in domains such as physics, mathematics, coding, and more. Other news has Sam Altman posted on X and says that his company’s first open model is late until the end of this summer.
Desperately seeking: People can ask the chatbot for answers, which can also be generated from news content shot without the knowledge of the publisher – there’s no need to click on the blue Google link. And that’s hurting the publisher.
nice? Mattel and Openai are working together to create “products powered by AI” whatever it is. As part of the transaction, Mattel employees can also access OpenAI tools such as ChatGpt Enterprise to “enhance product development and creative ideas.”
“Privacy Disaster”: Reporter Amanda Silverling has tried out the meta AI app and discovered that he is publishing people’s questions. “The meta doesn’t indicate what your privacy settings are or where you’re posting when you post. So is it searching for how to meet ‘Big Booty Women’ if you’re logged in to Meta AI on Instagram and your Instagram account is public,” she writes.
iPad for work: The iPads 26 brings new features to devices from 15 years old, making it possible to actually use it for a day’s work.
analysis
Recent waves of headlines and posts have raised questions about Blueskiing, from concerns about slowing growth to turning the platform into a left-leaning echo chamber, which users claim to be too serious. These criticisms capture part of the conversation, but do not reflect the full picture of what Bruski is working on. However, if not checked, these perceptions can pose a real challenge to the future growth of the platform.
Source link