OpenAI Acquires Sky

While we were still figuring our way around Open AI’s Atlas browser dropped earlier this week, the team pulled another move with the acquisition of Sky, an AI assistant interface for Mac. It is safe to say OpenAI’s objective is to have a full takeover of our daily mobile usage; posing a threat to Google products. Know who else is a threat? Samsung with its Galaxy XR release for almost half the price of Apple Vision Pro.

OpenAI just picked up Sky, a smart AI assistant for Mac that was being built by Software Applications, Inc. The product hadn’t launched publicly yet, but the idea behind it is pretty compelling, it’s an AI interface that floats over your desktop and can see what you’re doing on your screen, then jump in to help with writing, planning, coding, and other tasks. Think of it as an AI assistant that actually watches and understands your work, then takes action in your apps when you need it to.

The team behind Sky has some serious pedigree. Co-founders Ari Weinstein and Conrad Kramer are the same duo who created Workflow, which they sold to Apple back in 2017. That app eventually became what we now know as Shortcuts. They both stuck around at Apple for a few years before starting Software Applications in August 2023. The third co-founder, Kim Beverett, spent nearly a decade at Apple working on major products like Safari, Messages, and FaceTime.

Weinstein said they’ve always wanted computers to be more helpful and intuitive, and with today’s AI technology, they can finally make that happen. He’s excited about bringing that vision to hundreds of millions of people now that they’re joining OpenAI.

The acquisition makes a lot of sense for OpenAI as they push to get their technology into people’s daily routines, especially for the many businesses that run on Mac computers. But it’s also an interesting move given what’s happening with Apple. Apple has been playing catch-up on AI, and they’re expected to launch a much smarter version of Siri next year.

They’ve already rolled out other AI features, writing tools, live translation, image creation, under their Apple Intelligence umbrella, and they’re even partnering with OpenAI to hand off questions that Siri can’t handle to ChatGPT.

OpenAI didn’t reveal what they paid for Sky, but the company had raised $6.5 million from investors including OpenAI’s own CEO Sam Altman, Figma CEO Dylan Field, Context Ventures, and Stellation Capital. OpenAI noted that Altman’s investment came through a fund where he held a passive interest. The deal was led by Nick Turley, who heads up ChatGPT, and Fidji Simo, OpenAI’s CEO of Applications, and got the green light from OpenAI’s board.

Amazon is working on AI-powered smart glasses for its delivery drivers, and the goal is pretty straightforward, shave precious seconds off each delivery.

The glasses are designed to give drivers a truly hands-free experience. Right now, delivery drivers are constantly juggling their phones, packages, and keeping an eye on their surroundings. With these glasses, they’ll be able to scan packages, follow turn-by-turn walking directions, and capture proof of delivery all without touching their phone. The glasses use AI-powered sensors and computer vision along with cameras to create a heads-up display that shows delivery tasks and warns about potential hazards.

Here’s how it works: when a driver parks at a delivery spot, the glasses automatically turn on. They help the driver find the right package in their vehicle, then guide them to the exact delivery location. This is especially useful in tricky spots like apartment buildings with multiple units or sprawling business complexes. The glasses are paired with a small controller that clips onto the driver’s vest, which holds the controls, a swappable battery, and an emergency button.

Amazon clearly put some thought into practical concerns too, the glasses support prescription lenses and even have transitional lenses that adjust to changing light conditions automatically.

The company is testing the glasses with delivery drivers across North America right now and plans to fine-tune the technology before rolling them out more widely. But Amazon’s already thinking ahead about what’s next. Down the road, they want to add “real-time defect detection” that could alert drivers if they accidentally drop a package at the wrong address. The glasses will also be smart enough to spot pets in yards and adjust to hazards like low-light situations.

The smart glasses weren’t the only thing Amazon announced on Wednesday. They also showed off a new robotic arm called “Blue Jay” that can work alongside warehouse employees to pick and sort items from shelves. And they introduced an AI tool called Eluna that’s meant to provide operational insights at their warehouses.

General Motors is bringing Google Gemini to its vehicles starting next year, adding a conversational AI assistant across Buick, Chevrolet, Cadillac, and GMC brands. The announcement came Wednesday at GM’s Forward event in New York City.

GM joins other automakers embracing generative AI, Stellantis is working with French AI firm Mistral, Mercedes integrated ChatGPT, and Tesla brought Grok to its cars. For GM, the move makes sense since their vehicles already have “Google built-in” with Google Assistant and Maps.

Dave Richardson, GM’s senior vice president of software and services, says Gemini will be a big upgrade over current voice assistants. Today’s systems are frustrating because they need specific phrases and struggle with accents. Large language models don’t have these problems, they understand context, handle different ways of speaking, and feel more natural.

This means easier messaging, route planning with extra stops, and answering random questions like the history of a bridge you’re driving over. The assistant will roll out as an over-the-air update for OnStar-equipped vehicles from 2015 and up.

But GM’s thinking bigger. They want to build custom AI that taps into your car’s systems to send maintenance alerts, suggest routes, explain features, and even turn on your heat before you get in. They’ll train models specifically on vehicle specs and test options from multiple AI companies, not just Google.

Privacy is a key focus, especially after GM’s recent controversy over selling customer driving data to insurance companies. Richardson emphasized that drivers will control what the assistant can access, and any data collected improves the product rather than getting sold for revenue. GM has brought in new leadership, including Christina Montgomery from IBM’s privacy team, to handle data governance properly. “Everything that we’re going to do is going to be driven by customer consent,” Richardson said.

After months of rumors, Samsung finally unveiled its Galaxy XR headset to compete with Apple’s Vision Pro. It costs $1,800, nearly half the price of Apple’s offering. It’s available now in the U.S. and Korea.

The Galaxy XR runs on Google’s Android XR operating system and Qualcomm’s Snapdragon XR2+ Gen 2 platform. On paper, the specs are competitive, the micro OLED display packs 27 million pixels (six million more than Vision Pro) with a resolution of 3,552 x 3,840. The refresh rate is 90Hz compared to Vision Pro’s 120Hz. The headset also weighs 545 grams, making it noticeably lighter than the Vision Pro’s 750-800 grams.

Samsung says the lighter design comes from smart engineering, advanced materials and component optimization that balance comfort with durability. The frame distributes pressure across your forehead and back of your head to minimize discomfort during longer sessions. Battery life gives you about two hours of general use or two and a half hours of video playback.

The headset features two high-resolution pass-through cameras, six world-facing tracking cameras, and four eye-tracking cameras. In pass-through mode, you can see the real world around you and use hand gestures to circle objects and search for information about them.

Google’s ecosystem is baked in throughout. You can use Gemini to navigate places on Google Maps in immersive 3D, ask it to find YouTube videos and learn more about what you’re watching, and convert 2D photos into 3D in Google Photos. Apps like Google Maps, YouTube, Circle to Search, and Google Photos are all optimized for XR.

For productivity, you can multitask with multiple apps open at any size, browser, documents, music app, whatever you need. Entertainment options include streaming on the 4K micro OLED screen, with services like Crunchyroll, HBO Max, and Peacock already updating their apps for Android XR.

Early adopters get a solid deal. Anyone who buys the Galaxy XR by year’s end receives an Explorer Pack with 12 months of Google AI Pro, YouTube Premium, and Google Play Pass, plus trials of YouTube TV, NBA League Pass for the 2025-2026 season, and access to games like NFL PRO ERA and apps like Adobe’s Project Pulsar.

The timing is notable, Apple just launched an updated Vision Pro last week with an M5 chip that improves display rendering by 10% and bumps the refresh rate to 120Hz. It also added 30 minutes of battery life, giving it two and a half hours of general use and three hours of video playback. Samsung’s making its move in a market that’s still finding its footing, betting that a lower price point and Android’s flexibility will win over customers.

Captions is built for smaller scale operations where you don't have an editor and need to create the entire video directly in the app.