Meta just rolled out firmware v11 for Ray-Ban Meta glasses to early access users in the US and Canada, adding real-time AI video and continuous voice chat. Wearers can now ask questions about whatever is in front of them without repeating wake words, alongside live translation for Spanish, French, and Italian. Why it matters: It beats Google to getting multimodal live vision onto consumers' faces.
Meta CTO Andrew Bosworth just laid out the company's roadmap for 2025, calling smart glasses the first hardware category completely defined by AI. The company is rolling out an early access preview of Live AI this month, shifting assistants from prompt-based replies to proactive helpers that follow your day.
OpenAI just launched Projects for ChatGPT, letting users group related chats, custom context, and files into dedicated workspaces. The update integrates directly with Canvas and web search for multi-step tasks, rolling out now to Plus, Pro, and Teams subscribers before reaching Enterprise users next year.
Perplexity just rolled out custom web sources for Spaces, letting you lock searches down to specific domains and uploaded files. You can plug in target URLs via a new interface menu to cut through generic web noise. It essentially turns Perplexity into a focused research engine tuned strictly to your trusted sources.
Google just revamped NotebookLM with experimental Gemini 2.0 Flash integration, a new three-panel interface, and an interactive mode for its viral Audio Overviews that lets you jump into the conversation with your voice. The company is also rolling out NotebookLM Plus, a premium tier offering over 5x higher limits for heavy users and enterprises.
OpenAI just published a postmortem for Wednesday’s three-hour outage across ChatGPT, Sora, and its API. It turns out the culprit wasn’t a cyberattack, but an internal telemetry rollout that overloaded Kubernetes API servers and broke DNS resolution. The company is now revamping its deployment safeguards to prevent similar lockouts.
OpenAI just rolled out a festive 'Santa Mode' across all ChatGPT platforms for the holiday season. Users can tap the snowflake icon next to the chat bar to talk with a jolly-sounding Santa Claus, and the conversations won't save to history or touch your memory. It is free for everyone, but only until the end of the month.
Google just unveiled Android XR, a new operating system built with Samsung for headsets and smart glasses. The platform deeply bakes in Gemini so you can talk about what you see, multitask across virtual screens, and run standard Android apps. Samsung's first headset, Project Moohan, drops next year.
Anthropic just rolled out its Claude 3.5 Haiku model to Claude web and mobile users. The lightweight model matches or beats the previous flagship 3 Opus on certain benchmarks, excelling at coding, data extraction, and longer outputs. The only real trade-off is the lack of image analysis.
Anthropic just unveiled Clio, an automated analytics tool designed to study real-world Claude conversations without compromising user privacy. The system uses Claude to distill and cluster millions of chats into anonymized macro-topics, shielding personal details from human eyes while catching coordinated abuse. Why it matters: it gives AI labs a way to discover unknown safety risks without reading private messages.
Time just rolled out an AI chatbot alongside its 2024 Person of the Year feature, built in partnership with Scale AI. Readers can ask questions about the story, digest it into short summaries, translate it into six languages, or generate audio playback. It is a major experiment to see if conversational AI can keep digital readers engaged longer.
Grok just rolled out a faster, sharper version of Grok-2 to everyone on X for free. The upgraded model runs three times faster with better accuracy and multilingual support, while bundling web search, citations, and the Aurora image model. Developers also get updated API endpoints at lower token pricing.
Google just rolled out Deep Research for Gemini Advanced users, turning the assistant into an autonomous research partner. It creates a multi-step search plan, browses the web across several passes, and compiles findings into a clean report with source links. Plus, a speedy experimental preview of Gemini 2.0 Flash is now open to all Gemini users on the web.
Google just introduced its next-gen Gemini 2.0 model family, rolling out Gemini 2.0 Flash Experimental to developers and Gemini web users today. The new model beats 1.5 Pro at twice the speed while bringing native multimodal outputs, tool calling, and a new Deep Research mode. Why it matters: Google is moving fast from basic chatbots to proactive browser-controlling and code-executing agents.
Apple just released iOS 18.2, iPadOS 18.2, and macOS Sequoia 15.2, bringing the next wave of Apple Intelligence. Users can now create Genmoji, generate artwork in Image Playground, and tap into ChatGPT directly through Siri and Writing Tools. It is live now for localized English, with other languages rolling out starting this April.
Microsoft just started rolling out an updated Copilot experience to Windows testers, ditching the clunky web app wrapper for a floating quick view UI above the taskbar. It triggers via Alt + Space and stays pinned on top, though existing apps using the same shortcut might fight Copilot for priority.
OpenAI just graduated its Canvas workspace out of beta, making the collaborative interface available to all free and paid ChatGPT users. The update brings native Python code execution with live data visualizations, side-by-side diff editing, and integration with custom GPTs. It is rolling out on the web and Windows desktop apps right now.
Character.AI and Google are facing a fresh lawsuit in Texas after a 17-year-old alleged the platform's chatbots encouraged him to self-harm and alienated him from his parents. The suit claims defective product design and a lack of safety guardrails for minors. Google stated it has no involvement in Character.AI's tech, while the legal test for chatbot liability continues to heat up.
Solos just launched the AirGo Vision smart glasses starting at $299, putting multimodal ChatGPT vision and voice right on your face. The standout feature is a modular, swappable frame that lets you detach the camera entirely when entering privacy-sensitive spaces. It also plans to support Claude and Gemini alongside its 2,500-interaction battery life.
Microsoft AI CEO Mustafa Suleyman just pushed back against claims that AGI is achievable on current hardware like Nvidia GB200s, estimating it will take two to five more chip generations—or five to ten years. He also distanced practical AGI from sci-fi superintelligence, arguing builders should focus on useful, accountable AI companions instead of chasing the singularity.
