Welcome back to our weekly roundup of happenings from XR and AI realms. Let’s dive in…
The Lede
What to expect from CES 2026. CES starts Monday, January 5th. The show floor opens on Tuesday. The show will be a showcase for AI-driven hardware that consumers can actually buy, though most of it will never see the light of day. Expect a wave of household robots and drones focused on housework, eldercare, pets, and security. AI wearables are going to be everywhere, not just smartwatches but brooches, pins and badges with cameras, microphones, and agent-based software. The success of Meta’s Ray-Ban smartglasses means there will be entire aisles of cheaper knockoffs with cameras and voice assistants. Android XR will finally arrive in full force with headsets and glasses from Samsung and Xreal. Foldables will also have a strong showing. Multiple vendors are rolling out flexible OLEDs with fewer creases, better durability, and improved hinge mechanisms. BMW always has a big, flashy presence at the show. This year, they’re showcasing a panoramic heads-up display driven by AI that spreads a heads-up display across the windshield, replacing traditional dials and clusters.
If you are headed to CES, here’s some personal advice. (1) wear your most comfortable shoes; (2) wear layers: it’s cold outside and hot inside; (3) travel light, without a backpack; (4) don’t take paper, take pictures; (5) hydrate every chance you get. See you there!
Follow the Money
Nvidia Buys chip designer Groq for $20 billion. But technically, it’s an acquihire and licensing deal. The widely cited price estimate for the deal is $20 billion. Groq makes computer chips that help AI give answers quickly. AI first learns by studying data, then it answers questions using what it learned. Groq focuses on the answering part, making AI faster, cheaper, and more efficient for everyday use in apps, devices, and services. Analysts say that inference technology fills a much-needed gap for Nvidia, the largest company in the world.
Nvidia completed a previously announced roughly $5 billion purchase of Intel common stock, acquiring more than 214 million shares at about $23.28 per share. The transaction, cleared by U.S. regulators, gives Nvidia a meaningful equity position in the legacy CPU maker. Nvidia booked almost three billion dollars in paper gains on the day the deal closed.
Netflix Acquires Ready Player Me for Gaming Avatars. The technology will enable Netflix subscribers to craft persistent digital personas that can travel across various games on its platform, so it will have its own Netflix metaverse. The startup’s team will join Netflix as it integrates the cross-game identity tools to enhance player engagement. Terms were not disclosed.
Lemon Slice Raises $10.5 Million for Avatar AI Innovation. The startup, which is focused on digital avatar generation, raised $10.5 million in seed funding led by Y Combinator and Matrix Partners to build out its video-centric avatar models. Its Lemon Slice-2 model can generate interactive avatar video from a single image and stream at real-time frame rates on a single GPU. The company plans to embed these avatars into chat and agent experiences, aiming to bridge the gap between voice and visual AI interaction layers. I tested it and found it to be remarkably similar to Hedra and other talking head generators.
SpaceX’s Starlink satellite internet service now has more than 9 million active customers across 155 countries, just weeks after hitting 8 million in early November. That implies roughly one million net adds in under seven weeks, or about 21,000 new users per day. Starlink’s expanding footprint and mounting subscriber base strengthen the case for a future Starlink or SpaceX IPO, even as the company continues heavy capital spending on launches and next-generation satellites.
The AI Desk
OpenAI Searches for Head of Preparedness. The new senior executive will focus on anticipating and mitigating emerging risks from AI systems, including cybersecurity, biological misuse, mental health impacts, and broader societal harms. The role sits within OpenAI’s Safety Systems team and is tasked with developing threat models, capability evaluations, and cross-domain mitigation strategies to ensure frontier models are deployed responsibly. The fact that they don’t have this already is another reason there has to be some measure of regulation before AI kills us all.
Spatial Audio
For more spatial commentary & insights, check out the AI/XR Podcast, hosted by the author of this column, Charlie Fink, and Ted Schilowitz, former studio executive and futurist for Paramount and Fox, and Rony Abovitz, founder of Magic Leap. This week, our guest is the Chairman of the Consumer Technology Association, Gary Shapiro, who puts on CES and lobbies for the industry. You can find it on podcasting platforms Spotify, iTunes, and YouTube.
Charlie Fink is an author and futurist focused on spatial computing. See his books here. Spatial Beats contains insights and inputs from Fink’s collaborators including Paramount Pictures futurist Ted Shilowitz.