Home Blog Page 7

OpenAI Codex 2026: The New macOS App Turns AI into Your Coding Teammate

Kaus
-
0
OpenAI Codex 2026: The New macOS App Turns AI into Your Coding Teammate

I’ve been testing OpenAI Codex macOS app since it dropped earlier today. And honestly, it’s changing how I think about coding with AI. Instead of just getting suggestions for my next line of code. I’m now managing multiple AI agents that actually complete entire features while I work on other stuff.

The launch comes at an interesting time. OpenAI says usage has doubled since they rolled out the GPT-5.2-Codex model back in late 2025. Last month alone, over 1 million developers gave it a shot. What caught my attention? They’re offering temporary free access for ChatGPT Free and Go users. And if you’re already paying for a plan, your rate limits just doubled.

What separates Codex from tools like GitHub Copilot is the scope of what it handles. You’re not getting autocomplete here. You’re delegating actual work—building features, squashing bugs, reviewing pull requests. Everything runs in isolated cloud sandboxes. Which means you can experiment without worrying about breaking your local setup. This fits into the broader trend we’re seeing in 2026 where AI tools are becoming more autonomous and less hand-holdy.

What is OpenAI Codex and How Does It Work?

After spending a few hours with Codex. I can break down what it actually does versus the marketing speak. It runs on GPT-5.2-Codex which handles those long, tedious coding tasks that used to eat up entire afternoons. The difference between this and earlier models is pretty noticeable when you’re working on something that requires maintaining context across hundreds of lines of code.

When I’m coding in my terminal or IDE, Codex can navigate my entire repository. It edits files, runs tests and does it all in secure cloud environments that mirror my codebase. The multi-agent feature is where things get interesting. I can have one agent refactoring my backend while another updates the frontend components. They work in parallel, which cuts down project time significantly.

The code review functionality surprised me. I expected basic syntax checking but it actually understands what my code is trying to accomplish. You can set reviews to happen automatically or request them when you need a fresh perspective on something tricky. Integration with Slack, Linear and GitHub works smoothly. I get notifications when agents finish tasks or hit roadblocks.

Let me give you a practical example from this morning. I told Codex to add user authentication to a React project I’m working on. It mapped out the work, assigned different agents to the frontend login component and backend JWT handling, wrote everything, ran the test suite and created a pull request. I reviewed it over coffee, requested a couple tweaks and merged it. The whole process took maybe 45 minutes versus the half-day I’d normally spend on it.

The New macOS App Features and First Impressions

OpenAI Codex
image source- chatgpt

Before today, working with Codex meant switching between the command line, various IDE extensions and the web interface. The new macOS app centralizes everything. I’ve got a dashboard showing all active agents, can jump between projects without losing context and monitor long-running tasks that might take a couple hours to complete.

The multi-agent workflow genuinely speeds things up. A feature that would normally take me two or three days wrapped up in about six hours because I had multiple agents handling different components simultaneously. The sandbox security gives me peace of mind agents have limited write access and restricted network calls. So there’s minimal risk of accidentally pushing something catastrophic to production.

My main gripe? It’s Mac-only at launch. I split time between my MacBook and a Windows desktop. So I’m stuck using the web interface on half my setups. OpenAI confirmed a Windows version is in development, but no timeline yet. The interface feels a bit unpolished in spots nothing dealbreaking, just rough edges you’d expect from a day-one release. The doubled rate limits for paid plans and free trial access make this a low-risk time to experiment.

OpenAI Codex vs Claude Code vs Cursor

I’ve been using Claude Code and Cursor for the past few months. so naturally I wanted to see how Codex compares based on actual use cases.

Claude Code still has an edge on complex reasoning tasks. When I’m debugging something that requires understanding multiple interconnected systems. Claude tends to provide deeper analysis. The plugin ecosystem is also more mature. But Codex’s parallel agent execution is something Claude doesn’t really match. If I need multiple things happening simultaneously, Codex wins. Plus, since I already use ChatGPT and other OpenAI tools everything syncs up nicely.

Cursor is a completely different experience. It lives inside your IDE and gives you real-time feedback as you type. I can see diffs immediately and accept or reject changes on the fly. It’s perfect for that hands-on, I want to see every change as it happens workflow. Codex is better when I want to delegate an entire chunk of work and check back later. I’m not watching it code I’m assigning tasks and reviewing completed work.

My workflow now involves all three, honestly. I use Cursor for active coding sessions where I want constant feedback. Codex handles bigger features I can delegate. Claude Code comes in when I need to debug something particularly gnarly. The free Codex trial makes testing this combination easy without committing financially.

How to Get Started with OpenAI Codex

Setting up Codex took me less than five minutes. I went to openai.com/codex, downloaded the macOS app and logged in with my existing ChatGPT account. ChatGPT Plus runs $20 monthly. Though the temporary free access lets you test everything before deciding if it’s worth the subscription.

Start with something manageable for your first task. I began with Refactor this Python script for better performance just to see how it approached optimization. Once you understand its workflow, you can tackle bigger projects. My second task was adding dark mode to a landing page moderately complex but not mission-critical if something went wrong.

A few lessons I learned the hard way: always review the agent’s plan before it starts executing. I skipped this once and the agent took an approach I wouldn’t have chosen. Also, those pull requests Codex creates? Read through them carefully. The code is usually solid, but I’ve caught a few edge cases the AI missed. Keep using sandboxes for anything touching production. I made this a hard rule after reading about someone who didn’t and regretted it.

Don’t try replacing your entire development workflow immediately. I’m still using my local tools for most things. Codex handles specific tasks where parallel execution or cloud delegation makes sense. For solo developers and solopreneurs, this is like having a junior developer on your team. You’re still architecting and making the important decisions, but repetitive implementation work gets offloaded.

What’s Next for OpenAI Codex in 2026

After spending most of today with the Codex macOS app, I think we’re seeing a genuine shift in developer tools. This isn’t just better autocomplete. It’s AI that can take ownership of complete features. The combination of the new app and temporary free access means 2026 might be when agentic AI coding moves from experimental to standard practice.

I’m expecting OpenAI to release the Windows version within a few months based on demand I’m seeing in developer communities. More automation features are probably coming. I wouldn’t be surprised if they offer local deployment options for enterprise teams with security requirements. They’ve been responsive to feedback so far which suggests rapid iteration ahead.

If you’ve been curious about AI coding tools but haven’t taken the plunge, now’s a good time. Download Codex and test it on a side project before committing to your main work. What would you delegate first if you had an AI teammate handling the implementation? I’d love to hear what other developers are planning to build with this drop your thoughts in the comments.

Nubia M153 Doubao Review: ByteDance’s AI Phone That Sold Out in 24 Hours

Maya Kapoor
-
0
Nubia M153 Doubao Review: ByteDance’s AI Phone That Sold Out in 24 Hours

ByteDance just dropped their first smartphone and honestly? It’s causing quite a stir. The Nubia M153 Doubao flew off the shelves all 30,000 units gone within 24 hours of launch in China. At 3,499 yuan (roughly $480) this isn’t just another phone with AI features slapped on. It’s a collaboration with ZTE that actually does something different.

I’ve been tracking ByteDance’s moves for a while now from their Doubao AI competing with ChatGPT to how they’ve integrated AI across their ecosystem. But jumping into hardware? That’s a whole different game. After digging into the specs and capabilities, I can see why people are hyped.

What Makes the Doubao AI Different?

Nubia M153
image source- nubia official

Look, we’ve all used Siri , Google Assistant or Bixby. You ask them to set a timer or play music and they’re great. But the Doubao AI in this phone? It’s playing a completely different sport.

The feature they’re calling GUI proxy functionality basically lets the AI take control and do things for you. Not just open this app but actually navigate through multiple apps, compare prices, hunt down coupons and complete purchases. You literally just say find me the cheapest delivery option and order it and the AI goes to work. It’ll only ping you when it needs final confirmation.

I’m talking about real tasks too. Booking dinner reservations, editing your photos, scheduling hospital appointments, even telling your taxi driver to change routes mid-trip. This goes way beyond the usual voice command party tricks.

After testing countless AI tools and assistants over the past few years. This level of autonomy is genuinely new territory. It’s less like using a phone and more like having a personal assistant who actually knows how apps work.

The Hardware Stuff

Screen and Build

You’re getting a 6.78-inch LTPO display with 1264×2800 resolution. The adaptive refresh rate is smart. It adjusts based on what you’re doing to save battery. Size-wise. It’s 163mm tall, 77mm wide, 8.5mm thick and weighs 212g.

Having reviewed tons of phones. I’d say this hits that Goldilocks zone. Big enough that watching videos feels immersive. But not so massive you can’t use it with one hand when you’re standing on the train.

What’s Powering This Thing

Inside, there’s a Snapdragon 8 Gen 2 (the Premium Edition, which is basically the slightly juiced-up version) with 16GB of RAM and 512GB storage. That’s the same chip you’d find in flagship phones from last year that cost way more.

Here’s the thing about AI phones they need serious processing power. Running those AI models locally isn’t light work. So having this kind of horsepower makes sense. It’ll handle everything you throw at it. from gaming to video editing to those AI tasks running in the background.

Camera Setup Worth Talking About

This is where things get interesting. Four cameras all rocking 50MP sensors:

  • Main shooter: 50MP, big 1/1.3-inch sensor, optical stabilization, f/1.68 aperture (translation: excellent in low light)
  • Ultra-wide: 50MP, 12mm equivalent for those sweeping landscape shots
  • Telephoto: 50MP with 60mm reach and OIS for portraits and zooming without losing quality
  • Selfie cam: 50MP with autofocus (most front cameras don’t have that)

What stands out is the consistency. Most phones cheap out on the ultra-wide or telephoto, but Nubia went all-in across the board. Both the main and telephoto cameras have optical stabilization too. Which is clutch if you shoot video or photos in less-than-ideal conditions.

Battery Life and Charging

They crammed a 6000mAh battery in here. For perspective, most flagships sit around 4500-5000mAh. Translation? This thing will last.

Charging options are solid: 90W wired (crazy fast), 15W wireless and 5W reverse charging if you need to juice up your earbuds or someone else’s phone in a pinch.

As someone who’s constantly creating content and testing devices. Battery anxiety is real. A 6000mAh battery means I can actually work a full day without hunting for outlets.

The Extras

There’s NFC for payments, infrared (which is surprisingly handy for controlling TVs and AC units), ultrasonic fingerprint sensor under the screen, laser autofocus for the cameras, five microphones (overkill? Maybe, but great for voice recognition), dual speakers and USB-C with 3.2 Gen1 support.

The Bigger Picture

Here’s what’s wild—people were so eager to get this phone that resale prices shot up thousands of yuan over retail. That kind of demand isn’t just hype; it shows people are genuinely curious about what AI can do when it’s baked into a phone properly.

ByteDance says they’re planning another batch before the end of 2026. But no concrete dates yet. Having watched enough tech launches. I can tell you that this kind of reception usually means something’s resonating with people beyond just specs.

My Honest Take on Nubia M153

This phone represents something bigger than just another device launch. We’ve been hearing about AI phones for a while, but most have been disappointing. Just regular phones with some AI photo filters or predictive text. The M153 actually delivers on the promise of AI making your phone smarter and more capable.

Is it perfect? Probably not. Is it available everywhere? Definitely not. But does it show where things are headed? Absolutely.

For $480, you’re getting flagship performance, an impressive camera system. Marathon battery life and AI features that actually feel futuristic rather than gimmicky. The biggest hurdle is getting your hands on one.

If ByteDance can scale production and maybe expand beyond China, this could shake up expectations across the industry. After years of incremental upgradesslightly better cameras, marginally faster processors.It’s refreshing to see something that feels genuinely different.

Whether this becomes the mainstream standard or stays a niche curiosity depends on execution. But one thing’s clear: the era of truly intelligent smartphones isn’t coming it’s already here.

AISOMA: Google’s AI Tool That Teaches You Dance Moves

Arjun Patel
-
0
AISOMA: Google’s AI Tool That Teaches You Dance Moves

What if you could dance with an AI that’s learned from 25 years of professional choreography? That’s exactly what Google Arts & Culture created with AISOMA and honestly it’s one of the most interesting uses of artificial intelligence I’ve seen lately.

Unlike the usual AI tools that write emails or create images. AISOMA does something totally different. It watches you dance and then teaches you new moves based on a famous choreographer’s entire career archive. Yeah you read that right the AI actually responds to how you move.

What Is AISOMA?

AISOMA came from a partnership between Google Arts & Culture Lab and Wayne McGregor. Who’s basically a legend in contemporary dance. The name mixes “AI” with “soma” (Greek for body), which makes perfect sense once you try it.

The whole thing works like this: you dance for a few seconds in front of your webcam. The AI studies what you just did and throws back a suggestion for your next move, all inspired by McGregor’s choreographic style. Then you try that move and the AI responds with another one. It’s like having a back-and-forth conversation. Except you’re using your body instead of words.

How They Built It

McGregor’s been choreographing for 25 years and his team archived everything over 4 million different poses and movements. Google’s AI learned from all of that. So when you move in front of your camera, the system isn’t just randomly generating choreography. It’s drawing from decades of real creative work.

The pose detection happens through your webcam. The AI figures out what your body’s doing matches it against patterns it learned from McGregor’s archive and creates something new. It’s not copying old dances it’s making fresh combinations that still feel true to McGregor’s distinctive style.

From Private Tool to Public Experiment

Here’s what I find really cool about AISOMA’s history. Google built this back in 2019, but only McGregor and his professional dancers could use it. For six years it stayed inside his studio as a creative tool that helped professional dancers explore new movement ideas.

Dancers would perform something, check what AISOMA suggested, then reinterpret that suggestion in their own way. McGregor found it super valuable for breaking creative blocks and pushing his company in unexpected directions.

Then in 2025, Google updated everything for McGregor’s Infinite Bodies exhibition in London and opened it up to everyone. Now anybody can mess around with the same tool that professional dancers have been using for years. That’s a pretty big deal.

Actually Using the Thing

AISOMA
image source- google labs

You don’t need any fancy equipment or dance training. Just go to the AISOMA website and let it access your camera.

Move however you want. Dance, jump, wave your arms, whatever feels right. The AI watches and analyzes your movement in real-time.

Within seconds, you’ll see a visual representation of new choreography on your screen. It shows you the movements the AI is suggesting based on what you just did. Try performing what it showed you don’t worry about getting it perfect. Your interpretation becomes the next input.

That’s where it gets interesting. Your version of the AI’s suggestion generates another suggestion. Which you interpret again and the cycle continues. You’re basically co-creating with a machine that’s learned from one of the world’s best choreographers.

Why This Actually Matters

Most AI discussions focus on whether machines will replace creative jobs. AISOMA flips that script entirely. It’s designed to enhance human creativity, not substitute for it.

Think about the typical AI tools people use daily. They complete tasks for you—write your emails, summarize documents and generate marketing copy. AISOMA doesn’t work for you; it works with you. There’s a massive difference there.

Plus, it makes professional-level choreographic knowledge accessible to regular people. Before AISOMA, learning from Wayne McGregor meant expensive workshops or getting into elite dance programs. Now? Anyone with internet access can engage with his creative approach from their bedroom.

There’s something else worth mentioning. McGregor’s archive isn’t just sitting in storage somewhere. It’s active and interactive, constantly participating in new creative work with people all over the world. That’s a fascinating way to think about preserving artistic legacy.

Who Can Actually Use This?

Don’t assume this is only for trained dancers. I’ve seen all kinds of people get value from AISOMA.

Fitness people use it to discover new movement patterns for their routines. Dance students explore concepts they’d never encounter in regular classes. Teachers demonstrate how technology and art can intersect in unexpected ways. Some folks who’ve never danced a day in their lives try it just for fun and end up hooked.

There’s zero barrier to entry. No subscription fee, no software download, no prerequisites. You just need a webcam and enough curiosity to give it a shot.

Where Creative AI Is Headed

AISOMA shows us something important about AI’s future in creative fields. These tools work best when they collaborate with humans rather than trying to replace them.

The most exciting applications aren’t about automation. They’re about expansion helping us break our usual patterns, suggesting directions we wouldn’t think of ourselves and making specialized knowledge more accessible.

Google proved that AI’s creative potential goes way beyond text and image generation. Physical movement and dance are now part of the equation. Which opens up tons of possibilities we’re only starting to explore.

If you’re curious, head over to Google Arts & Culture and search for AISOMA. Give it a try. Worst case scenario, you’ll spend five minutes dancing awkwardly in front of your laptop. Best case? You might discover a whole new way to think about creativity and movement.

Microsoft’s Maia 200 AI Chip: The Battle Against Nvidia Just Got Real

Ethan Cole
-
0
Microsoft’s Maia 200 AI Chip: The Battle Against Nvidia Just Got Real

Remember when Microsoft relied entirely on Nvidia for its AI computing power? Those days are officially over. This week, Microsoft launched the Maia 200. Its second-generation AI chip and it’s making some pretty bold claims about taking on the giants of the AI hardware world.

The chip went live in a data center in Iowa on Monday with another location planned for Arizona. Microsoft says this isn’t just another incremental upgrade. It’s a serious attempt to reduce its dependence on Nvidia. Which currently controls about 85% of the AI chip market.

What Makes the Maia 200 Different

The Maia 200 is specifically built for AI inference. Which is the process of actually running AI models after they’ve been trained. Think of it this way: training an AI model is like teaching someone how to ride a bike. While inference is them actually riding it every single day. As AI companies grow, inference costs have become a massive part of their expenses. Which is exactly why chips like this matter so much.

Microsoft claims the Maia 200 delivers impressive performance numbers. It hits 10 petaflops in FP4 precision and around 5 petaflops in FP8 performance. That’s four times faster than Amazon’s Trainium 3 chip in FP4 workloads. The chip packs over 140 billion transistors and comes with 216GB of HBM3e memory with 7 TB/s of bandwidth.

Built on TSMC’s 3-nanometer process, the Maia 200 runs at 750W. That’s almost half the power draw of Nvidia’s Blackwell B300 Ultra chip. Which uses 1,400W. Microsoft says this makes the Maia 200 about 30% more efficient per dollar compared to the first-generation Maia 100.

Breaking Nvidia’s Software Stronghold

Here’s where things get really interesting. Microsoft isn’t just competing on hardware. It’s going after Nvidia’s biggest advantage: CUDA, the programming platform that keeps developers locked into Nvidia’s ecosystem.

To challenge this, Microsoft is offering Triton an open-source programming language that was developed with major contributions from OpenAI back in 2021. Triton lets developers write GPU code in a Python-like language without needing years of CUDA expertise. OpenAI says researchers with zero CUDA experience can use Triton to write highly efficient GPU code that matches what expert programmers produce.

This is a big deal. Switching costs have kept many developers tied to Nvidia for years. If Triton works as advertised it could make it much easier for companies to move their AI workloads to alternative chips like the Maia 200.

The Chip Inside

Maia 200
image source- microsoft

The Maia 200 includes some clever design choices borrowed from emerging AI chip companies. Microsoft packed it with 272MB of on-die SRAM, a type of super-fast memory that gives speed advantages for chatbots and AI systems handling lots of simultaneous user requests. This approach mirrors strategies used by companies like Cerebras Systems. Which recently signed a $10 billion deal with OpenAI and grok. which licensed its inference technology to Nvidia in a non-exclusive deal.

One Maia 200 node can run today’s largest AI models with room to spare for even bigger models coming in the future. The chip is designed to handle rapid responses during demand spikes while staying within tight power limits that data centers increasingly face.

Why This Matters Now

Microsoft isn’t alone in this race. Google has been drawing interest from major Nvidia customers like Meta. Which is actively working to close software gaps between Google’s TPU chips and Nvidia’s offerings. Amazon has its Trainium line and Apple is reportedly working on its own AI chips too.

The AI chip market is expected to reach around $2 trillion by early next decade. With Nvidia holding such a dominant position, every major cloud provider is investing heavily in custom silicon to control costs and differentiate their services.

For Microsoft specifically, this move makes strategic sense given its deep partnership with OpenAI. The company needs massive amounts of computing power to run ChatGPT and other AI services. Reducing dependency on external chip suppliers could save billions over time.

What Happens Next

The Maia 200 will first power Microsoft’s own Azure cloud infrastructure. The company hasn’t announced when regular Azure customers will be able to rent servers powered by these chips but developers can already start using the control software.

Microsoft faced some delays getting here. Design changes requested by OpenAI and staff turnover pushed mass production into 2026. But now that the chip is live and processing real workloads. We’ll soon see whether Microsoft’s performance claims hold up in production environments.

For anyone watching the AI industry, the Maia 200 represents more than just another chip launch. It’s a clear signal that the era of Nvidia’s near-total dominance might be starting to shift. Whether Microsoft can actually deliver on its promises remains to be seen, but one thing is certain: the competition for AI computing power just got a whole lot more interesting.

Google Disco: Turn Your Browser Tabs Into Custom Apps With AI

Sophia Lin
-
0
Google Disco: Turn Your Browser Tabs Into Custom Apps With AI

Google has launched an exciting new experiment called Disco that could change how we browse the web. Instead of juggling dozens of tabs while researching or planning something online. Disco uses AI to transform those messy tabs into clean interactive apps that help you get things done.

What Is Google Disco?

Disco is Google’s latest experiment from Google Labs designed to test new ideas for the future of web browsing. The main feature being tested right now is called GenTabs. Which stands out as a truly innovative approach to managing online tasks. Think of it as your personal assistant that watches what you’re doing online and creates helpful tools automatically.

Understanding GenTabs

GenTabs is the star of the show. This feature turns your open browser tabs into custom. Interactive web applications tailored to whatever you’re trying to accomplish. The best part? You don’t need to write a single line of code or have any technical skills.

Here’s what makes it special: GenTabs uses Gemini 3 Google’s most intelligent AI model to understand what you’re working on by analyzing your open tabs and chat history. It then stitches all that scattered information together into one focused, useful app.

How Does It Work?

The process is surprisingly simple. As you browse the web with multiple tabs open. GenTabs proactively watches what you’re looking at and suggests interactive apps that might help you. For example, if you have several tabs open about different travel destinations. It might suggest creating a trip planner app.

You can also tell GenTabs what you need using plain everyday language. Just describe the tool you want and the AI builds it for you. Want to refine it? Simply chat with it like you’re talking to a friend and it adjusts the app based on your feedback.

The AI pulls content from your open tabs. Your chat history and can even grab additional relevant information from the web automatically. Everything it creates links back to the original sources. So you can always verify information.

What Can You Create?

Google has shared several examples of what GenTabs can build:

  • Interactive trip planners with calendars, timelines and maps
  • Meal planning apps
  • Garden planning tools
  • Solar system explorers with 3D visuals
  • Custom study helpers
  • Activity comparison tools for tourists

The possibilities are endless because each GenTab is generated based on your specific tabs and your specific goal. If you can imagine it GenTabs can probably build it.

Real-World Example

Let’s say you’re planning a trip to Japan. You’ve got tabs open for flights, hotels, tourist attractions and restaurants. Instead of switching between all those tabs. GenTabs creates an interactive app with a zoomable map a calendar and information organized in neat sections. It might even show you crowd levels at different tourist spots and suggest the best times to visit.

Key Benefits

GenTabs offers several advantages for anyone who spends time online:

  • Reduces tab clutter by combining multiple tabs into one organized app
  • Saves research time by automatically gathering and organizing information
  • Keeps you focused on your goal instead of getting distracted
  • Works without coding so anyone can use it regardless of technical background
  • Stays goal-oriented by creating task-specific applications

How To Get Access

Google Disco
image source- google labs

There’s a catch: Disco isn’t available to everyone yet. Google is currently running a waitlist for people who want to test it. You’ll need to sign up on the Google Labs website and it’s initially only available on macOS.

Unlike other Google gemini 3 features you won’t find GenTabs in the regular Chrome browser. You’ll need to download and use Disco. Which is a separate application designed specifically for this experiment.

The Future of Web Browsing

Google has made it clear that GenTabs is just the first feature being tested in Disco. The company plans to introduce more features over time as they experiment with new ways to browse the web. If the ideas developed through Disco prove successful. They might eventually appear in larger Google products like Chrome.

This experiment represents a major shift from passive browsing to active. AI-assisted web navigation. Instead of you doing all the work to organize and make sense of information. The AI handles the heavy lifting while you focus on making decisions and taking action.

Is It Worth Trying?

If you’re someone who regularly juggles multiple tabs for research, planning, or complex projects. Disco could be a game-changer. The ability to turn scattered information into organized, interactive apps without any coding makes it accessible and practical for everyday users.

Google Disco shows us where web browsing might be headed: a future where your browser doesn’t just display information but actively helps you accomplish your goals.

Yahoo Scout 2026: The Tech Giant’s AI-Powered Return to Search

Sophia Lin
-
0
Yahoo Scout 2026: The Tech Giant’s AI-Powered Return to Search

Yahoo has officially re-entered the search engine race with Scout. A new AI-powered answer engine that launched in beta on January 27, 2026. Available now at scout.yahoo.com and through the Yahoo Search app on iOS and Android. Scout represents Yahoo’s first proprietary search technology in over 15 years, marking a significant shift after the company outsourced search to Microsoft Bing back in 2009.

Scout combines Anthropic’s Claude AI model with Yahoo’s 30 years of data and Microsoft’s Bing grounding API to deliver conversational search results that prioritize accuracy and source transparency. CEO Jim Lanzone described the launch as an opportunity to supercharge the original Yahoo mission of being the trusted guide to the internet.

What Makes Yahoo Scout Different?

Yahoo Scout
image source- yahoo.com

Unlike competitors such as ChatGPT and Perplexity. Scout takes a publisher-first approach designed to drive traffic back to content creators. Each response features prominent inline citations with bright blue highlights that reveal sources when users hover over them. Along with dedicated featured source sections that encourage clicks to original publishers.

The first iterations of AI engines did not nearly enough to send traffic downstream. Lanzone told Search Engine Land. This philosophy extends to Yahoo joining Microsoft’s Publisher Content Marketplace pilot. An initiative aimed at supporting sustainable revenue streams for publishers and content creators.

The platform includes interactive digital media, structured lists and tables and visible source links aimed at making answers easier to verify. In early testing, The Verge found Scout provided more accurate answers and featured nine links on a single results page compared to competitors that obscure links behind icons or faint buttons.

Powered by Anthropic’s Claude

Yahoo Scout runs on Anthropic’s Claude AI model. One of the top foundational models in the market. Anthropic, founded in 2021 by former OpenAI members including siblings Daniela and Dario Amodei is backed by major tech companies like Amazon and Google.

However, Yahoo extensively customizes the Claude model by integrating its proprietary datasets. Creating a unique user experience unlike generic Anthropic deployments. When you’re serving hundreds of millions of users. You need AI that can do more than retrieve information. It has to reason, synthesize and explain, said Ami Vora, Head of Product at Anthropic.

Eric Feng, Senior Vice President and General Manager of Yahoo Research Group and former founding CTO at Hulu led the development effort. Yahoo’s deep knowledge base 30 years in the making allows us to deliver guidance that our users can trust, Feng said.

Scout Intelligence Platform Across Yahoo Properties

Yahoo Scout
image source- yahoo.com

Beyond standalone search, Yahoo is deploying Scout capabilities across its entire ecosystem through the Scout Intelligence Platform. The integration includes:

Yahoo Mail provides email summaries and actionable item extraction. Such as automatically adding calendar events. Yahoo Finance offers one-click stock analysis with company financials, analyst ratings and real-time stock move explanations. Yahoo News delivers article highlights and daily digest audio summaries. Yahoo Sports features game breakdowns and key moment highlights. Yahoo Shopping includes product insights and shoppable links.

This embedded approach transforms Scout from a simple search tool into an AI companion that enhances user experiences across Yahoo’s network.

Monetization Strategy

Yahoo plans to monetize Scout through Microsoft Advertising-powered CPC ads appearing at the bottom of some responses and affiliate commissions on commerce-related queries. The platform will remain free for all users contrasting sharply with OpenAI’s subscription-dependent model for ChatGPT.

Competing in a Crowded AI Search Market

Yahoo enters the AI search space as the third-largest search engine in the United States, boasting approximately 250 million U.S. users and over 500 million user profiles globally. The company processes 18 trillion consumer signals annually across its properties.

Still, competition is fierce. Google and OpenAI dominate the AI search landscape. While Perplexity has established itself as a search-first AI tool with persistent citations and real-time information access. ChatGPT, while primarily generation-focused, excels in depth and reasoning for complex queries.

Yahoo’s advantage lies in its publisher-friendly approach and massive existing user base. By emphasizing traffic generation for content creators and leveraging decades of user data. Scout positions itself as a more ethical and sustainable alternative in the AI search ecosystem.

What’s Next for Yahoo Scout?

Yahoo says the answer engine behind Scout will become more personalized over time. focusing on deeper experiences as it learns from user interactions. The beta launch represents just the beginning of Yahoo’s AI transformation, with the company clearly betting that its legacy infrastructure and publisher partnerships will differentiate Scout in an increasingly crowded market.

For users tired of AI search engines that obscure sources or fail to credit original content. Yahoo Scout offers a refreshing alternative. It aims to restore the social contract between search platforms and the publishers who create the web’s content.

Clawdbot to Moltbot: This Open-Source AI Assistant Went Viral

Kaus
-
0
Clawdbot to Moltbot: This Open-Source AI Assistant Went Viral

A new AI assistant has taken the tech world by storm over the past few days and it’s not from any of the usual suspects like Google, Apple or Microsoft. This is an open-source project that exploded from a niche developer tool into a full-blown phenomenon complete with Mac Mini buying sprees, serious security concerns and a forced name change that happened just yesterday.

The assistant is called Moltbot. If that name sounds unfamiliar you probably know it better as Clawdbot. Which is what everyone called it until Anthropic stepped in with trademark concerns. The rebrand happened fast but the buzz around this tool hasn’t slowed down.

What Is Clawdbot?

Clawdbot
image source- moltbot

Moltbot (Clawdbot) is an AI assistant that runs on your own computer instead of living in some company’s cloud. You can connect it to basically all your messaging apps at once like WhatsApp, Telegram, Discord, Slack, Signal and iMessage. Message it from any platform and it remembers your entire conversation history because everything gets stored as simple text files on your machine.

The real difference is that Moltbot actually does things instead of just talking. Need to check your calendar? It pulls up your schedule. Want to send an email or run code on your computer? It handles that. One user called it a junior system administrator who never sleeps. Which honestly captures what makes this different from asking Siri to set a timer.

You can tell it to monitor your inbox and automatically schedule meetings or have it check your calendar each morning and send you a briefing without you asking. The system is agentic, meaning it takes action on its own rather than waiting for constant instructions.

The Sudden Mac Mini Craze

Clawdbot
image source- apple.com

Something unexpected happened as word spread about Moltbot. People started buying Apple Mac Minis like crazy. The reason? You need a computer running all day to keep Moltbot active and Mac Minis hit the sweet spot of being small, quiet, energy efficient and reasonably priced.

Logan Kilpatrick, a product manager at Google DeepMind, tweeted that he ordered a Mac mini as he joined the rush. Google searches for Mac Mini spiked over four days. One developer even posted screenshots showing 12 Mac Minis being configured at the same time. That’s a shopping spree worth more than $7,000 just for running AI assistants.

It reminds me of when cryptocurrency mining made graphics cards impossible to find except now it’s AI tools driving the hardware shortage instead of Bitcoin.

How Fast Did This Thing Blow Up?

Peter Steinberger built Moltbot as a personal project. He’s an Austrian developer who founded the document software company PSPDFKit which is now called Nutrient. He wanted something to manage his calendar and smart home without sending all his data to corporate servers.

The GitHub repository sat relatively quiet for months with around 10,000 stars. Then last weekend everything changed. The project jumped to nearly 30,000 stars within days. The Discord community swelled past 8,900 members. Over 156 people started actively contributing code.

Nobody’s entirely sure what triggered the explosion. Probably a combination of AI hype hitting critical mass, frustration with locked down corporate assistants and genuine curiosity about what an open-source AI helper could actually accomplish. The community has already built over 100 ready to use skills that anyone can plug into their setup.

Security Researchers Found Major Problems

As Moltbot gained popularity security researchers started investigating. What they discovered was troubling.

SlowMist is a blockchain security firm that found more than 900 Moltbot instances running online without any password protection. These weren’t harmless test servers. They were actively leaking private data. Anthropic API keys that cost real money, Telegram tokens, Slack credentials and months of personal chat histories were just sitting exposed for anyone to grab.

Security researcher Jamieson O’Reilly publicly flagged the issue and warned that hundreds of API keys and private conversations were at risk. The Moltbot documentation now includes urgent warnings about enabling password authentication and proper security configuration before running anything.

This reveals the tradeoff with open-source tools. You get complete control and transparency but security becomes your responsibility. Corporate assistants handle this automatically while potentially scanning your data for their own purposes. With Moltbot protecting your information is entirely on you.

The Name Change Nobody Saw Coming

Yesterday Anthropic contacted Steinberger about the project’s name. The issue was that Clawdbot and its assistant persona Clawd were too similar to Claude ai which is Anthropic’s flagship AI product.

This creates an ironic situation since most Moltbot users actually run Claude as their underlying AI model. The project exists partly because Claude excels at complex reasoning and multi-step tasks. But Anthropic has invested millions building their brand and having a viral third party tool with a nearly identical name obviously creates confusion.

Steinberger and his team moved fast. Within hours they rolled out completely new branding. Clawdbot became Moltbot. Clawd became Molty. The Twitter handle switched to @moltbot and a fresh domain went live at molt.bot.

The team posted that Anthropic asked them to change the name because of trademark stuff but honestly Molt fits perfectly because it’s what lobsters do to grow. They kept their lobster mascot too. The project has maintained a crustacean theme from the beginning which somehow makes this whole saga more entertaining.

The transition wasn’t entirely smooth. Scammers immediately hijacked the old Twitter handle to promote fake cryptocurrency tokens. A bogus CLAWD token briefly hit an $8.48 million market cap before crashing. Steinberger had to coordinate with Twitter and GitHub to recover control of the abandoned accounts.

Why This Feels Different From Siri or Alexa

Clawdbot
image source- apple.com

Traditional voice assistants are locked down by design. You can only do what the company permits. Your data lives on their servers and they control what integrations exist.

Moltbot reverses that model completely. Everything runs on hardware you own. The code is open-source so anyone can inspect it, modify it or add features. If you want new functionality you or the community can just build it instead of waiting for Apple or Amazon to approve your feature request.

What This Means for AI Assistants

Moltbot’s viral moment reveals something about where we are with AI technology right now. People clearly want assistants that feel genuinely useful rather than glorified search engines. They want tools that integrate with their actual workflows instead of separate apps they need to remember to check.

There’s also growing interest in self hosted and privacy focused alternatives to big tech platforms. Running your own AI on your own hardware means nobody’s scanning your conversations for advertising or training future models on your personal data.

The security issues show this isn’t consumer ready yet. You need comfort with command line tools, server configuration and basic security practices. This remains very much a power user playground.

Is Moltbot Worth Trying?

For developers, tinkerers or early adopters comfortable with technical configuration. Moltbot is genuinely impressive. The ability to connect multiple messaging platforms while actually controlling your computer makes it feel like the AI assistant promised years ago.

But if you’re expecting an out of the box experience like Alexa this isn’t it. You’ll need to bring your own API keys which cost money. You’ll need to secure your setup properly or risk exposing your data. And you’ll probably need to troubleshoot various issues.

The community is active and helpful but this is early stage software experiencing explosive growth. Features break. Documentation falls behind. Security best practices are still being established.

The team posted during the rebrand that their mission stays the same. AI that actually does things. That core vision remains unchanged despite the new name.

Whether Molty catches on as quickly as Clawd did is uncertain. Name recognition matters especially for viral projects. But the fundamental appeal hasn’t changed. A powerful self hosted AI assistant you completely control. For people willing to invest the setup time it’s the most capable personal AI assistant available right now assuming you don’t accidentally leave it exposed to the internet.

Microsoft Teams Up with Mercedes F1 in Massive $60M Deal

Arjun Patel
-
0
Microsoft Teams Up with Mercedes F1 in Massive $60M Deal

Microsoft just made waves in the Formula 1 world with a partnership announcement that’s turning heads. They’ve signed a multi-year deal with the Mercedes-AMG PETRONAS F1 Team worth roughly $60 million per season kicking off in 2026. It’s a big move Microsoft’s jumping ship from Alpine to Mercedes right when F1’s about to undergo some of its most dramatic technical changes in decades.

The 2026 season isn’t just another year on the calendar. We’re talking brand new chassis, completely redesigned power units and stricter fuel regulations all aimed at making the sport greener and more efficient. Mercedes needed a serious tech partner to tackle these challenges and Microsoft clearly saw an opportunity.

It’s Not Just About Slapping Logos Everywhere

Mercedes F1
image source- microsoft.com

Yeah, the Microsoft logo will be plastered on Mercedes new W17 car. You’ll spot it on the front wings and airbox, plus on George Russell and rookie Kimi Antonelli’s racing suits. But honestly, the visual stuff is just scratching the surface.

What’s really happening here is that Microsoft’s Azure cloud platform and AI tools are getting baked right into Mercedes operations. We’re talking factory floor to race track. Judson Althoff, who runs Microsoft’s commercial business, summed it up nicely they’re putting their tech at the heart of racing performance, where milliseconds matter. In F1, that’s not marketing speak it’s the truth.

Drowning in Data

Get this: every Mercedes F1 car runs about 400 sensors that pump out 1.1 million data points per second. Yeah, per second. During a two-hour race, that adds up to an absolutely staggering amount of information. Azure’s job is to make sense of all that noise in real time, helping engineers spot patterns, adjust strategies on the fly and squeeze out every bit of performance.

Mercedes is already playing around with virtual sensors powered by Azure. Instead of waiting weeks to build and install physical hardware for testing. They can simulate scenarios in the cloud and get answers fast. In a sport where teams are constantly hitting up against budget caps and technical regulations, that kind of speed is massive.

Cloud Computing That Actually Makes Sense

Here’s something pretty clever: Mercedes uses Azure Kubernetes Service to scale their computing power up or down depending on what they need. Running heavy simulations before a race weekend? Crank everything up. Between races with lighter workloads? Dial it back and save money. It’s way more flexible than traditional server setups and it helps them stay within F1’s tight cost controls.

They’re also rolling out GitHub across their engineering teams to improve how everyone collaborates and shares code. Might sound boring, but when you’re racing against the clock to develop new parts and updates, workflow improvements matter.

The Boss Is Pumped

Toto Wolff, who runs the Mercedes F1 team was pretty straightforward about his excitement: “We’re delighted to partner with Microsoft, one of the world’s foremost technology leaders. By putting Microsoft’s technology at the center of how we operate. We’ll create faster insights, smarter collaboration and new ways of working.

You can tell he sees this as more than a sponsorship check. It’s a genuine competitive advantage.

Why Both Sides Are Winning

For Microsoft, this deal gets them prime real estate in front of 800 million F1 fans worldwide. The sport’s blown up in recent years especially in the U.S. so the brand exposure is huge.

For Mercedes, it’s all about staying ahead of the pack. They’ve been on a roll signing major partners lately. They just announced a PepsiCo deal in Decemberand stacking up tech advantages heading into 2026.

As F1 enters what might be its biggest shakeup in years. Mercedes and Microsoft are making a bet that cloud computing and AI will matter just as much as aerodynamics and engine power. Considering how data-driven modern racing has become, they’re probably right. It’ll be fascinating to see how this plays out once the lights go out next season.

FlashLabs Releases Chroma 1.0: Open-Source Voice AI That Responds in 135ms

Ethan Cole
-
0
FlashLabs Releases Chroma 1.0: Open-Source Voice AI That Responds in 135ms

The race to build faster, more natural voice AI just took an unexpected turn. While tech giants like OpenAI have been perfecting their closed-source voice models. San Francisco-based FlashLabs just dropped something the AI community has been craving: a fully open-source, real-time voice AI that actually works.

Released on January 22, 2026, Chroma 1.0 isn’t just another text-to-speech model wrapped in fancy marketing. It’s the first complete speech-to-speech system that operates natively in voice. Cutting out the traditional pipeline that makes most voice assistants feel sluggish and robotic.

What Makes Chroma Different from Other Voice AI Models

Chroma 1.0
image source- github.com

Most voice AI systems you interact with today follow a clunky three-step process: they convert your speech to text, process that text through a language model and then convert the response back to speech. It works but it’s slow and loses the natural flow of human conversation.

Chroma takes a completely different approach. It processes speech directly into speech, maintaining the emotional tone, pacing and natural rhythm that makes conversations feel human. The result? An end-to-end response time of just 135 milliseconds with SGLang optimization or about 147ms in standard configuration.

To put that in perspective, human conversational turn-taking typically happens around 200ms. Chroma operates well within that natural window, making interactions feel genuinely real-time rather than the delayed back-and-forth we’ve grown accustomed to with voice assistants.

The model also achieves a Real-Time Factor of 0.43. Which means it generates speech 2.3 times faster than playback speed. This efficiency ensures smooth streaming even during extended conversations, without the awkward pauses that plague traditional systems.

Voice Cloning That Actually Sounds Like You

Beyond speed, Chroma’s standout feature is its voice cloning capability. Give it just a few seconds of reference audio and it can generate a personalized voice that maintains consistency across multiple conversation turns.

FlashLabs reports a speaker similarity score of 0.817 in internal evaluations that’s nearly 11% better than the human baseline of 0.73. While voice cloning isn’t new, integrating it into a real-time dialogue system at this quality level is a genuine breakthrough.

This opens doors for applications that previously required expensive custom voice work. Think AI call centers that can speak in a company founder’s voice, gaming NPCs with unique personalities that persist across gameplay sessions or accessibility tools that help speech-impaired users communicate in their own voice.

Technical Architecture: Compact But Powerful

Chroma runs on a surprisingly lean 4-billion-parameter architecture, making it efficient enough for edge deployment rather than requiring massive cloud infrastructure. The system consists of three main components working together.

The Qwen-based Reasoner handles speech understanding and generates the initial audio tokens with a time-to-first-token of 119.12ms. A 1-billion-parameter LLaMA-style Backbone then produces audio hidden states in just 8.48ms. Finally the Decoder generates the remaining acoustic features across seven codebooks. Taking an average of 17.56ms per frame before the Codec Decoder reconstructs the final waveform.

For voice cloning, Chroma uses CSM-1B to encode reference audio into embeddings that condition the generation model. This architecture keeps computational requirements reasonable while maintaining quality a single H200 GPU can generate a 38.80-second response in just 16.58 seconds.

How Chroma Performs in Real-World Tests

FlashLabs evaluated Chroma on URO Bench, a standard benchmark for voice dialogue systems. Despite its compact size, the model achieved a 57.44% overall task accomplishment score on the basic track. It also posted competitive results on reasoning benchmarks like TruthfulQA and GSM8K, showing that reducing latency didn’t come at the cost of intelligence.

The latency breakdown reveals why Chroma feels so responsive. The Reasoner’s 119ms time-to-first-token represents the bulk of the initial delay. While the Backbone adds less than 10ms. Compare this to traditional voice AI platforms where end-to-end latency can range from 465ms in optimal conditions to over 950ms for some commercial solutions.

Real-World Applications Beyond Virtual Assistants

Chroma 1.0
image source- freepik.com

FlashLabs envisions Chroma powering a range of applications where natural voice interaction matters. The most obvious is customer service AI call centers could handle complex queries with voices that sound genuinely helpful rather than robotic.

Real-time translation is another compelling use case. Instead of the stilted, sentence-by-sentence approach current translation apps use. Chroma could enable fluid conversations between people speaking different languages, preserving tone and emotional context.

The gaming industry could benefit significantly too. Instead of recording thousands of voice lines for NPCs, developers could use Chroma to generate dynamic dialogue that responds naturally to player choices while maintaining character consistency. Healthcare applications could restore voices for patients who’ve lost speech capability, giving them back a crucial part of their identity.

FlashLabs is already deploying Chroma within its FlashAI voice agent platform. which focuses on transforming sales and customer experience through AI automation.

Availability and the Open-Source Advantage

Unlike OpenAI’s Realtime API. which remains firmly behind closed doors. Chroma 1.0 is fully open-source. FlashLabs released both the model weights and source code on Hugging Face and GitHub. Where it quickly climbed to the top of the multimodal category rankings.

This matters beyond just philosophical arguments about open versus closed AI. Developers can inspect exactly how the model works, fine-tune it for specific use cases and deploy it on their own infrastructure without ongoing API costs. For enterprises concerned about data privacy, running Chroma locally means sensitive conversations never leave their servers.

The open release also accelerates innovation. Researchers can build on FlashLabs work rather than starting from scratch, potentially leading to improvements that benefit everyone.

What This Means for Voice AI in 2026

Yi Shi FlashLabs founder and CEO, framed the release in ambitious terms: Voice is the most universal interface in the world. Yet it has remained closed, fragmented and delayed. With Chroma, we’re open-sourcing real-time voice intelligence so builders, researchers and companies can create AI systems that truly work at human speed.

That vision feels within reach now. Chroma demonstrates that you don’t need proprietary infrastructure or massive parameter counts to achieve natural voice interaction. The 4-billion-parameter architecture proves that efficiency and quality can coexist.

For developers frustrated by API rate limits and costs, Chroma offers a viable alternative. For researchers exploring new voice AI architectures. It provides a solid baseline to build upon. And for users tired of clunky voice assistants. It hints at a future where talking to AI feels as natural as talking to another person.

The voice AI landscape shifted this week and it happened in the open.

Frequently Asked Questions

What is FlashLabs Chroma 1.0?

Chroma 1.0 is the first fully open-source end-to-end real-time speech-to-speech AI model released by FlashLabs. featuring voice cloning and sub-150ms response latency.

How fast is Chroma’s response time?

Chroma achieves end-to-end time-to-first-token of approximately 135ms with SGLang optimization. Or 147ms in standard configuration, operating well within human conversational timing.

Can Chroma clone voices from short audio clips?

Yes, Chroma can generate personalized voices fromOjust a few seconds of reference audio, achieving a speaker similarity score of 0.817 nearly 11% better than human baseline performance.

Is Chroma 1.0 free to use?

Yes, Chroma is fully open-source with both model weights and source code available on Hugging Face and GitHub at no cost.

Logitech Rally AI Camera: New Smart Conference Cameras for Hybrid Teams

Liam Hayes
-
0
Logitech Rally AI Camera: New Smart Conference Cameras for Hybrid Teams

Logitech announced the Rally AI Camera and Pro on January 22, 2026. Its most advanced conference cameras designed for hybrid workspaces. Both models feature RightSight 2 AI technology that automatically frames meeting participants and includes room intelligence for occupancy tracking. The cameras offer 4K video quality with a 115-degree field of view and optional in-wall mounting.

Quick Specs:

  • Launch date: January 22, 2026
  • Models: Rally AI Camera ($2,499), Rally AI Camera Pro ($2,999)
  • Key feature: RightSight 2 AI framing
  • Video: 4K with 1-inch sensor
  • Field of view: 115 degrees
  • Available: Spring 2026 (Pro), Summer 2026 (standard)

Remote workers know the struggle. You’re on a video call and everyone in the conference room looks tiny or gets cut off when they stand up. Meanwhile, people in the office dominate the conversation because they’re right there. Logitech’s new Rally AI cameras tackle this problem head-on with smart framing that keeps everyone visible.

What’s New in the Rally AI Camera Series

Logitech released two models this week. The Rally AI Camera costs $2,499 and ships in summer 2026. It’s built for medium-sized meeting rooms with 4× digital zoom and works great for teams of 5-10 people.

The Rally AI Camera Pro is the bigger sibling at $2,999, launching in spring 2026. This one has a dual-camera system with 15× hybrid zoom. Which means it can capture details across large boardrooms and training spaces. Think rooms with 15-20 people where a single camera can’t get everyone clearly.

Both cameras use a 1-inch imaging sensor that handles low light well. You get 4K video quality and a 115-degree view that covers most conference rooms without fish-eye distortion. They come in graphite and off-white colors to match different office designs.

The standout design feature? You can mount the Rally AI Camera inside your wall. It’s the first conference camera that offers true in-wall installation, so it basically disappears when you’re not using it. The Pro model is half as deep as older Rally cameras despite packing more tech inside.

How RightSight 2 AI Framing Works

RightSight 2 is the brain behind these cameras. It watches your meeting and switches between three viewing modes automatically. Group View shows everyone at once. Speaker View zooms in on whoever’s talking. Grid View splits the screen into individual tiles so remote workers see each person clearly.

The AI runs directly on the camera, not in the cloud. This keeps things smooth because there’s no lag from sending video back and forth to servers. The camera tracks people separately from framing decisions, so movements look natural instead of jerky.

Here’s a practical feature: Camera Zone lets you mark where meetings actually happen. If your conference room has glass walls and people walk by outside, the camera won’t accidentally frame them. You draw a boundary and RightSight 2 ignores everything outside it.

The Pro model adds Presenter View, which launches after the initial release. This mode follows speakers as they walk around, perfect for training sessions or presentations where people don’t stay seated. The dual-camera system tracks movement across bigger spaces without losing focus.

Both cameras work with Zoom Intelligent Director and Microsoft Teams for multi-camera setups. If you have multiple Rally cameras in one room, the system coordinates them to create different angles automatically.

Room Intelligence Beyond Video Calls

Rally AI Camera
image source- logitech

These cameras do more than just record meetings. They detect when rooms are occupied or sitting empty. That data flows into Logitech Sync, the company’s management platform.

The occupancy tracking solves real problems. If someone grabs a conference room for an impromptu meeting, the camera notices and books the space automatically. When meetings end early it releases the room so other teams can use it. Companies waste a lot of time dealing with ghost meetings where rooms show as booked but nobody’s actually there.

People counting gives facility managers useful information. They can see which conference rooms get heavy use and which ones sit empty most days. That helps with decisions about office layout and real estate costs. Maybe that underused 12-person boardroom could become two smaller collaboration spaces instead.

Rally AI Cameras are designed to power the hybrid-first office. where the tech fades into the background to let the digital and physical worlds blend, said Henry Levak. He is Vice President of Product at Logitech for Business. From small walls to town halls, they provide a cinematic experience for meeting attendees while quietly solving problems that IT managers, Facilities teams and Workplace Experience professionals face every day.

Setup and Compatibility

IT teams can install these cameras with a USB connection or run them through a single Cat cable using Logitech’s Extension Kit. They work with Rally speakers for audio. Or you can connect professional systems from Shure, Q-SYS, Biamp and Nureva.

Remote management happens through Logitech Sync. IT staff can push firmware updates, check camera status and troubleshoot issues without visiting each conference room. The cameras have a physical privacy shutter that clearly shows when they’re off.

Each camera uses low-carbon aluminum in its construction and ships in packaging made from FSC-certified forests. Wall, ceiling and display mounting options give flexibility for different room layouts.

Rally AI Camera vs Rally AI Camera Pro: Key Differences

Rally AI Camera
image source- logitech.com
FeatureRally AI CameraRally AI Camera Pro
Price$2,499$2,999
Zoom4× digital15× hybrid
Camera SystemSingleDual
Launch DateSummer 2026Spring 2026
Best ForMedium rooms (5-10 people)Large spaces (15-20 people)
Presenter ViewNoYes (coming after launch)

Logitech Rally AI Camera Price and Availability

The Rally AI Camera Pro launches first in spring 2026 for $2,999. The standard Rally AI Camera follows in summer 2026 at $2,499. Both will be available through Logitech’s website and authorized resellers in graphite and off-white colors.

Holly Zhou, Logitech’s Head of Product Marketing for Team Workspaces explained that customers have been asking for an update to the original Rally Camera. People have loved this camera for its incredible optics, but we have heard time and time again that they want an update. They want updated experiences for now, the 2020s”.

Why This Matters for Hybrid Work

Hybrid work isn’t going anywhere. Companies need tools that make remote participants feel as present as people in the office. The Rally AI cameras address the visibility gap that frustrates both groups.

The in-wall mounting option changes how conference rooms look. Instead of having cameras and tech equipment dominate the space, everything can blend into the design. Rooms feel less like IT labs and more like actual meeting spaces.

Room intelligence features help companies use office space better. When you know which rooms get used and which ones sit empty. You can make smarter decisions about real estate. That’s real money saved, especially for companies paying premium prices for downtown office locations.

These cameras compete with conference systems from Poly, Cisco and Microsoft in the enterprise market. The pricing puts them in the professional category, not the consumer webcam space. Companies investing in hybrid work setups now have another option that combines video quality with AI smarts and facility management features.

Frequently Asked Questions

What is the Logitech Rally AI Camera?

The Logitech Rally AI Camera is a 4K conference camera with AI-powered auto-framing and occupancy detection, designed for hybrid meeting spaces. It uses RightSight 2 technology to automatically adjust framing based on who’s speaking and where people are in the room.

How much does the Rally AI Camera Pro cost?

The Rally AI Camera Pro costs $2,999 and launches in spring 2026. The standard Rally AI Camera is priced at $2,499 and arrives in summer 2026.

What is RightSight 2 technology?

RightSight 2 is Logitech’s AI framing system that automatically switches between group shots, individual speakers and grid layouts during meetings. It runs on the camera itself for smooth transitions without cloud processing delays.

When will the Logitech Rally AI cameras be available?

The Rally AI Camera Pro launches in spring 2026, while the standard Rally AI Camera arrives in summer 2026. Both models can be ordered through Logitech’s website and authorized business resellers.

1...678...22Page 7 of 22