Google co-founder Larry Page is quietly building Dynatomics, a stealth startup focused on applying AI to physical manufacturing. Led by former Kittyhawk CTO Chris Anderson, the team is developing AI that creates optimized object designs for direct factory fabrication. Why it matters: One of tech's richest minds is now betting AI can reshape physical industrial design.
The U.S. Department of Labor is investigating $13.8B data-labeling giant Scale AI for potential violations of the Fair Labor Standards Act, TechCrunch reported. The probe, active since at least August 2024, centers on whether the startup misclassified workers as independent contractors to bypass overtime and benefits. Scale AI denied wrongdoing, saying its platform provides flexible earning opportunities.
Johns Hopkins researchers just developed a biomimetic prosthetic hand blending rigid 3D-printed bones with soft silicone joints. Multi-layered fingertip sensors allow it to recognize everyday objects with 99.69% accuracy and hold a full water cup without crushing it. The prototype still requires an air compressor and awaits clinical amputee trials.
OpenAI just updated its ChatGPT macOS desktop app to edit code directly inside developer IDEs. The feature is available today for Plus, Pro, and Team subscribers, with free and enterprise access coming soon. It cuts out the annoying copy-paste routine, though early testers note you should keep your Cmd+Z muscle ready.
Tavus just launched three AI models—Phoenix-3, Raven-0, and Sparrow-0—designed to make video avatars feel human. The trio tackles micro-expressions, visual perception, and conversational turn-taking, cutting awkward dialogue pauses. They are now available via API, backed by a demo agent named Charlie that talks, sees, and searches.
Hedra just unveiled Character-3, an omnimodal AI model that handles image, text, and audio all at once, alongside its Hedra Studio platform. Instead of chaining separate tools to stitch lip-sync and character motion, creators can now generate expressive, synchronized video in a single unified workflow.
Mistral AI just released Mistral OCR, a dedicated API built to parse messy documents, math equations, and tables into structured markdown. It can process up to 2,000 pages per minute at $1 per 1,000 pages, and is already powering document understanding inside Le Chat.
Andreessen Horowitz just dropped its fourth Top 100 Gen AI Consumer Apps report, showing massive market shifts. ChatGPT doubled its weekly active users to 400 million in under six months, propelled by GPT-4o and o1, while newcomer DeepSeek stormed the web traffic ranks to claim #2 globally within days of launch. Vibe coding tools and AI video are also rapidly cementing their place in daily consumer habits.
Krea AI just plugged in a set of new video features from Pika Labs. Creators can now edit specific video regions, drop in AI-generated elements, smooth out motion with frame interpolation, and blend multiple images directly inside the platform.
Convergence just introduced Template Hub, a centralized library packed with hundreds of ready-to-deploy agentic workflows created by its users. With one click, the Proxy agent gets straight to work, and creators can share setups and track run counts. It basically turns complex daily automations into shared templates anyone can run.
McDonald's is rolling out a sweeping AI overhaul across its 43,000 restaurants worldwide. Powered by edge computing from Google Cloud, the system uses connected sensors for predictive kitchen maintenance—including its notoriously broken McFlurry machines—and smart drive-thrus to ease crew stress.
DuckDuckGo just pushed its AI chat platform Duck.ai out of beta and upgraded AI search summaries. Users can access models like GPT-4o mini and Claude 3 Haiku for free and without an account, routing prompts through privacy proxies with local-only storage. Best of all, DDG added settings allowing users to dial down or completely turn off AI features.
Tencent Hunyuan just open-sourced HunyuanVideo-I2V, releasing weights, inference code, and LoRA training scripts. The 13B parameter model turns a single image into a 5-second 2K clip complete with background audio, and even handles motion driving and lip-syncing. It is available on GitHub and HuggingFace, with a web demo open to the public.
Google just expanded its medical AI system AMIE from one-off diagnoses to ongoing disease management across multi-visit consultations. Powered by a dual-agent Gemini setup grounded in clinical guidelines, AMIE matched or outperformed primary care doctors across 100 simulated patient cases. It marks a crucial step toward AI-assisted chronic care, though real-world clinical validation is still pending.
Anthropic just overhauled its Console into a unified workspace for building and iterating on AI apps. Teams can now share and collaborate on prompt libraries directly in the console instead of copy-pasting across docs. It also adds full support for Claude 3.7 Sonnet, letting developers tweak thinking token budgets and export production-ready code in one click.
Butterfly Effect's new autonomous agent Manus just went viral, with invite codes fetching thousands of dollars after claims it outperforms OpenAI's tools. But beneath the hype and multi-model routing (including Claude and Qwen), early testers find it frequently crashes, hits endless loops, and fails basic tasks like ordering food or booking flights. The team says it's still in closed beta.
Prime Video just started experimenting with AI-assisted dubbing for select licensed movies and TV shows. The initial test covers 12 titles previously lacking dubbed tracks, offering English and Latin American Spanish with human localization pros reviewing the output for quality control.
Qwen just launched QwQ, a mid-sized reasoning model that matches heavyweight frontier systems like DeepSeek-R1. Alibaba found that scaling reinforcement learning yields massive leaps in math and coding, proving mid-sized models can punch way above their weight against massive MoE giants.
Perplexity just rolled out a new voice mode to its macOS desktop app. Mac users can now trigger hands-free search anytime by hitting ⌘ + ⇧ + M. The update is live right now on the Mac App Store.
OpenAI just began rolling out GPT-4.5, its largest model to date, to ChatGPT Plus subscribers over the next one to three days. It promises richer world knowledge, fewer hallucinations, and better conversational flair, though Sam Altman openly warned it is not a dedicated reasoning model. It also comes with a hefty price tag—roughly 15 times more expensive per token than GPT-4o.
D-ID just partnered with Microsoft to bring its real-time AI avatars directly to Azure. Developers and enterprises can now integrate interactive digital humans into Microsoft Teams and other apps, offering lip-synced multi-language translation and sign language accessibility features.
Google just integrated Gemini 2.0 into AI Overviews in the U.S. to tackle coding and advanced math, while dropping the login requirement and expanding access to teens. The search giant also launched 'AI Mode' in Labs, a deep-reasoning search experience that runs concurrent multi-topic queries and supports conversational follow-ups. Early access is currently rolling out to Google One AI Premium subscribers.
Apple just rolled out AI-powered review summaries in the latest iOS 18.4 beta 2, condensing common user feedback directly on App Store listings. The summaries appear above user reviews, update at least weekly, and are currently live for eligible English apps in the US before expanding later this year.
LTX Studio just rolled out LTXV 0.9.5, an upgraded release of its open-source AI video model. The new version officially adds commercial licensing, keyframe conditioning, and native ComfyUI support, alongside higher resolution and fewer artifacts in longer video sequences.
Stability AI just scored a strategic investment and partnership with advertising titan WPP. The deal will plug Stability’s multimodal models for image, video, 3D, and audio directly into WPP’s AI operating system, WPP Open. Why it matters: It gives Stability a massive enterprise testing ground and validates its commercial push under new leadership.
OpenAI is reportedly preparing to launch specialized AI agents with monthly subscriptions ranging from $2,000 to $20,000 for tasks up to PhD-level research. The rollout features three tiers covering business, development, and research, with SoftBank committing $3 billion. OpenAI is betting enterprises will pay senior executive wages for automated expertise.
Reinforcement learning creators Andrew Barto and Richard Sutton just won the Turing Award and immediately used the moment to slam AI safety practices. Speaking to the FT, they warned that commercial hype is pushing unvetted models onto millions of users, likening current releases to building a bridge and testing it with real traffic.
Google just rolled out a suite of AI features for Google Shopping, led by "vision match," which turns text descriptions of your dream outfit into AI images and matches them with real buyable clothes. The update also brings Gemini-powered full-face AR makeup try-ons and expands virtual clothing try-on to pants and skirts. Why it matters: Google is directly bridging the gap between a vague idea in your head and an actual checkout cart.
Codeium just dropped Wave 4 for its Windsurf Editor, aimed straight at making vibe coding smoother. The standout Previews feature lets developers click components in web previews and auto-pipe console errors directly into the Cascade agent. It also adds tab-to-import and auto-fixes linter bugs for free. Basically, it kills the endless copy-paste loop of web dev.
Apple just unveiled the new MacBook Air powered by the M4 chip, now starting at $999 with 16GB of unified memory by default. The upgraded Neural Engine accelerates on-device AI tasks up to 3x faster than the M1 model, built specifically to power Apple Intelligence. Pre-orders are live today, with shipments arriving March 12.
Apple just introduced the updated Mac Studio powered by M4 Max and the new M3 Ultra chips. The compact desktop supports up to 512GB of unified memory and Thunderbolt 5, allowing developers to run LLMs with over 600 billion parameters entirely in memory. It basically turns a desktop Mac into a serious local AI powerhouse.
Uber and Waymo just launched autonomous rides in Austin, Texas, giving Uber riders a chance to get picked up by an all-electric Jaguar I-PACE with no human driver behind the wheel. The ride comes at no extra cost, and passengers can easily opt in or out. It plants robotaxis right in Tesla's home territory ahead of its own planned rollout.
TechBullion just published a breakdown of the controversial Jenny Mod for Minecraft Java Edition. The adult-focused mod introduces a virtual companion with customizable features, special in-game skills, and AI-driven interactions. Why it matters: While designed strictly for mature solo players, it shows the growing appetite for interactive companion NPCs in sandbox gaming.
xAI just rolled out its Voice Mode to all iOS users on the Grok app for free, removing the need for a paid subscription. The update brings natural conversational AI straight to any iOS device running the official app. It is a clear push to grab daily voice users from rival assistants.
A federal judge just shot down Elon Musk's request for a preliminary injunction to block OpenAI's shift into a for-profit company, ruling the plaintiffs failed to meet the bar for extraordinary relief. The decision keeps OpenAI's restructuring on track for now, though Musk's broader legal claims against the startup and Microsoft will still head to trial.
The Los Angeles Times is now deploying AI to flag opinionated pieces with a "Voices" tag and auto-generate bulleted "Insights" offering alternative viewpoints at the bottom of articles. The unvetted automated summaries are already sparking backlash from its journalists' union and producing clumsy, awkward takes.
Google just added an on-device AI scam detector to Google Messages to catch conversational text scams that start out friendly before turning nasty. The AI analyzes SMS and RCS chats from unsaved contacts in real time without sending your private messages to the cloud. It is enabled by default and rolling out first to English users in the US, UK, and Canada.
Microsoft just started rolling out a native version of its Copilot Windows app to Windows Insiders, nearly a year after turning it into a basic web wrapper. The updated app borrows design cues from the macOS version, complete with an improved UI and a new side panel. It is currently rolling out via the Microsoft Store, with a wider release expected in the coming weeks.
Deutsche Telekom just partnered with ElevenLabs to embed its Gen FM audio tech directly into the Magenta AI app, letting users turn news articles into podcasts in seconds. The telecom giant also invested in ElevenLabs’ Series C round. It signals a major shift toward making AI speech a default smartphone interface.
ElevenLabs just integrated Anthropic's Claude 3.7 Sonnet into its Conversational AI platform. Developers can now combine the hybrid-reasoning model with realistic voice synthesis, building voice agents that handle nuanced queries with fewer hallucinations and tighter context tracking.
Apple just refreshed the iPad Air with its M3 chip, making the midrange tablet fully ready for Apple Intelligence. It packs a 60% faster Neural Engine over the M1 model to speed up on-device AI tasks, landing in 11-inch and 13-inch sizes at the same $599 starting price alongside a cheaper Magic Keyboard.
Infinity Sim Wings just launched the first AI-enabled Airbus A320 simulator at its training facility near Gatwick Airport. The setup adds uninterrupted AI air traffic control coverage alongside automated ground and cabin crew interactions. The goal is to expose pilots to real-world operational workloads and evaluate cockpit resource management without leaving the safety of a flight sim.
OpenAI just launched NextGenAI, pledging $50M in research grants, compute funding, and API access across 15 leading institutions including MIT, Harvard, and Oxford. The initiative aims to fast-track discoveries in healthcare, metascience, and rare manuscript digitization. It is OpenAI's biggest academic push yet to embed its tech inside top labs.
UC San Diego’s Hao AI Lab just pitted top AI models against Super Mario Bros. using a custom framework. Anthropic’s Claude 3.7 took the top spot, while reasoning heavyweights like OpenAI’s o1 flunked out because taking several seconds to "think" in a platformer means instant death.
TSMC just committed at least $100 billion to expand its US semiconductor manufacturing footprint, funding two new factories in Phoenix alongside its existing three. CEO C.C. Wei confirmed the new sites will build advanced AI chips. It is a massive bet to onshore supply chains ahead of planned semiconductor tariffs.
California state Senator Scott Wiener just introduced SB 53, a fresh AI bill reviving the least controversial parts of the vetoed SB 1047. It shields whistleblowers at frontier AI labs from retaliation if they report catastrophic risks, while proposing a public cloud computing cluster called CalCompute for researchers.
Google just integrated its Data Science Agent directly into Colab, powered by the Gemini 2.0 model family. Users can upload CSV, JSON, or text files up to 1GB and ask plain-English questions to clean messy datasets, build visualizations, or write SQL. It rolled out for free this week under Colab's standard compute limits.
Google just rolled out three AI initiatives to combat biodiversity loss. The company is launching its first nature-focused startup accelerator across the Americas, committing $3M in grants to Brazilian nonprofits, and open-sourcing SpeciesNet—its computer vision model that identifies animal species from camera trap photos. Why it matters: It hands field researchers and developers scalable conservation tools for free.
T-Mobile parent Deutsche Telekom just announced an "AI Phone" built with Perplexity, putting Perplexity Assistant directly on the lock screen to eliminate app-hopping. Slated for later this year under $1,000, the device also integrates tools from ElevenLabs and Google Cloud AI. It marks Perplexity's push to evolve from an answer engine into an action machine.
Stability AI just teamed up with Arm to run its text-to-audio model, Stable Audio Open, directly on smartphone CPUs without an internet connection. Powered by Arm's KleidiAI libraries, generating an 11-second audio clip dropped from 240 seconds to under 8 seconds—a 30x speedup for on-device creators.
Opera just previewed Browser Operator, an agentic AI built directly into the browser to handle routine web tasks like shopping and bookings. Instead of cloud VMs or taking screenshots of your screen, it runs locally and reads the page DOM directly, keeping your login data safe on-device while letting you take the wheel anytime. It is slated to launch soon via Opera's AI Feature Drops.
A fully driverless Maserati MC20 Coupe just hit 197.7 mph (318 km/h) at the Kennedy Space Center, shattering the autonomous land speed record. Powered by algorithms from Politecnico di Milano and Indy Autonomous Challenge, the team pushed the AI software to its absolute limit to stress-test systems and eventually improve high-speed highway safety.
Google just showcased new Gemini multimodal features at MWC 2025, allowing users to ask questions using real-time video feeds or whatever is displayed on their phone screen. You can now film surroundings or browse shopping apps while getting instant AI feedback. The features roll out to Android Gemini Advanced subscribers later this month.
Anthropic just secured $3.5 billion in Series E funding led by Lightspeed Venture Partners, rocketing its valuation to $61.5 billion. The massive war chest will fund compute infrastructure, next-generation frontier models, and AI alignment research right on the heels of Claude 3.7 Sonnet's launch. It firmly cements Anthropic as a titan capable of matching OpenAI toe-to-toe in enterprise adoption.
Microsoft just unveiled Dragon Copilot, a unified voice AI assistant for healthcare that combines Dragon Medical One's dictation with DAX Copilot's ambient listening. Clinicians can draft notes in real time, automate clinical orders, and search trusted medical sources, saving roughly five minutes per encounter. It arrives in the U.S. and Canada this May.
Meta researchers just unveiled Avat3r, an AI system that generates an animatable, photorealistic 3D avatar head in minutes using only four phone selfies. It uses a vision transformer to predict 3D Gaussians, drastically lowering the hardware barrier from specialized multi-camera rigs. While rendering currently crawls at 8 FPS on an RTX 3090, it points toward effortless avatar setup for future headsets.
Apple is reportedly facing major AI rollout delays, with a ChatGPT-like conversational Siri pushed back to iOS 20 at best. Internal data also indicates that real-world Apple Intelligence usage is 'extremely low', leaving iOS 19 with few significant AI upgrades.
The world’s largest call center operator Teleperformance is deploying Sanas AI to neutralize the accents of its Indian customer support reps with zero latency. Management insists turning accents into neutral English cuts handling times and boosts satisfaction, but investors weren't convinced by the $104 million AI roadmap, dragging its stock down 11%.
Google just brought new Gemini capabilities to all Workspace business users in Sheets, letting the AI surface correlations, flag outliers, and generate advanced visuals like heatmaps. Instead of wrestling with spreadsheet formulas, users can simply tap the spark icon and ask plain-English questions to analyze their data.
OpenAI just teamed up with nine U.S. national labs to host a massive AI Jam Session, putting its frontier models into the hands of over 1,000 researchers across institutions like Oak Ridge and Berkeley. Scientists are stress-testing the models on complex domain-specific problems, with findings set to shape future research-focused AI systems in an upcoming report.
OpenAI just launched its AI video generation tool Sora in the EU and the UK. The rollout officially brings the popular video model to users and creators across both markets.
Amazon CEO Andy Jassy revealed that a brand-new lineup of "beautiful" devices is landing this fall to accompany the upgraded Alexa Plus. The conversational assistant costs $19.99 a month or comes bundled with Prime, and Amazon's hardware leadership hinted that richer screens will play a massive role in the next generation.
Meta is reportedly planning to launch a dedicated Meta AI mobile app during the second quarter, taking direct aim at OpenAI's ChatGPT. The company also intends to test a paid subscription tier. While Meta AI already reaches hundreds of millions inside Facebook and WhatsApp, a standalone app could unlock deeper personalization and hardware sync.
Microsoft just rolled out major upgrades to Azure AI Foundry, giving enterprise customers preview access to OpenAI's new GPT-4.5. The platform also introduced new model distillation workflows, private virtual network integration for secure agent deployments, and Magma, an architecture built to coordinate hundreds of AI agents in parallel.
Figure just moved up its home robotics roadmap by two years, with plans to begin alpha testing its Figure 02 humanoid in living rooms later in 2025. CEO Brett Adcock credited Helix, the startup’s new generalist VLA model, for the rapid shift. Why it matters: While rivals stick to factory floors, Figure is taking on the messy real world surprisingly early.
Sesame just came out of stealth with a remarkably expressive voice AI and plans for wearable companion glasses. Backed by top venture firms and led by Oculus co-founder Brendan Iribe, the assistant handles natural interruptions and banter with ease, with open-source models supporting 20+ languages slated for the coming months.
Ideogram just rolled out Ideogram 2a, its fastest and most cost-effective image model yet, generating photorealistic graphics and typography in seconds. The update cuts pricing by 50% compared to version 2.0 across both the web app and API. Ideogram also introduced public inspiration feeds that let anyone browse top community prompts without logging in.
Pika just launched Pika 2.2, bringing 10-second video generation and crisp 1080p resolution to its platform. The update introduces Pikaframes, giving users control over key frame transitions anywhere between 1 and 10 seconds for more complex visual storytelling.
DJI just unveiled the Dock 3, a rugged drone-in-a-box system that lets drones launch directly from moving vehicles with an optional gimbal mount. Built for the Matrice 4D and 4TD, it packs an IP56 rating, handles extreme temperatures from -30 to 50 degrees Celsius, and starts at $21,059. Why it matters: Mobile launches allow inspection and emergency teams to deploy aerial scans without ever having to stop the truck.
Mystic just introduced its new Structure Reference tool, giving creators precise control over structural integrity during image generation. It lets users preserve composition across drastically different styles—from cartoons to photorealism—aimed at film, 3D, and architectural concept workflows.
Meta just unveiled Aria Gen 2, its next-generation smart glasses built for machine perception and robotics research. The 75-gram frames run SLAM and multi-modal tracking entirely on-device via custom silicon, packing new nosepad sensors to isolate the wearer's voice and pulse. Why it matters: It provides AI researchers with a wearable testbed to collect egocentric data for training embodied AI and assistive tech.
You.com just launched ARI, a professional-grade research agent designed for enterprises that digests up to 400 web sources simultaneously. Unlike standard agents that read pages sequentially, ARI delivers a cited report with interactive charts and a downloadable PDF in about five minutes. It is currently available as a preview on You.com Labs.
AWS just unveiled Ocelot, a prototype quantum computing chip that cuts error correction costs by up to 90%. Built around 'cat qubits' that naturally suppress environmental noise, the microchip integrates correction hardware from the ground up. AWS researchers say the architecture could speed up the timeline to practical fault-tolerant quantum computing by up to five years.
OpenAI just released a research preview of GPT-4.5, its largest and most capable chat model to date. Built by scaling unsupervised learning and compute on Azure supercomputers, the model delivers broader world knowledge, reduced hallucinations, and better conversational nuance. ChatGPT Pro users and API developers can test it right now.
Nvidia CEO Jensen Huang just pushed back on fears of slowing AI demand, revealing that next-gen reasoning models require 100 times more computation than when ChatGPT debuted. Addressing market concerns after DeepSeek's rapid rise, Huang argued that step-by-step reasoning actually multiplies the need for chips. Why it matters: Despite worries that efficiency gains would cut spending, Nvidia insists smarter AI only makes the appetite for hardware grow faster.
Microsoft just introduced Phi-4-multimodal and Phi-4-mini, a pair of lightweight models built for edge hardware and Copilot+ PCs. The 5.6B multimodal variant natively ingests speech, vision, and text in a single architecture, topping open speech recognition leaderboards while staying compact enough to run smoothly on-device.
Inception Labs just launched Mercury Coder, the first commercial-scale diffusion language model that generates entire code sequences in parallel rather than predicting tokens sequentially. Running on standard Nvidia H100 GPUs, it hits over 1,000 tokens per second while matching models like GPT-4o Mini. Why it matters: It proves text can borrow from image generation to slash AI inference latency and costs tenfold without specialized hardware.
Mark Engineering just launched the Mark 1, a $129 smart bookmark made of Grade 5 titanium designed to track your reading. The device features a small screen and a rotary knob that syncs reading progress to a companion app, generating summaries, key quotes, and reading statistics. It's a fresh attempt to digitize paper books, though detailed specs and potential subscription plans remain unconfirmed.
Microsoft just rolled out its Copilot app to Mac computers, allowing users to summon the assistant over any app via the Option-Space shortcut. It includes Copilot Daily for podcast-style news roundups alongside voice chat that gracefully handles mid-sentence interruptions and corrections.
Amazon just introduced Alexa+, a major generative AI overhaul of its smart assistant powered by Amazon Bedrock. It can autonomously orchestrate across thousands of third-party apps to book appointments, order food, or manage smart homes with full context memory. Why it matters: It shifts Alexa from a simple voice remote into a proactive personal agent, and Prime members get it for free ($19.99/month otherwise).
Donald Trump just sparked widespread outrage after sharing an AI-generated video on Truth Social reimagining war-torn Gaza as a flashy luxury resort. The clip features a topless Benjamin Netanyahu, Elon Musk dancing under falling dollar bills, and a massive golden Trump statue, drawing fierce blowback even from die-hard supporters who called it tasteless.
ElevenLabs just rolled out Scribe, its first speech-to-text model designed to tackle real-world audio in 99 languages. It delivers structured JSON outputs with word-level timestamps, speaker diarization, and audio event tags like laughter, outperforming Whisper Large V3 on benchmark tests. It is available now via API and the web dashboard, with a low-latency real-time edition coming soon.
Qualcomm just unveiled Dragonwing, a dedicated brand targeting industrial IoT, networking, and enterprise infrastructure. The lineup includes the IQ series packing 100 INT8 TOPS for rugged environments and the Q series optimized for edge AI and Wi-Fi 7. Qualcomm is keeping Snapdragon for consumer gadgets while sending Dragonwing to challenge Nvidia and AMD in factories.
IBM just open-sourced Granite 3.2 under the Apache 2.0 license, bringing experimental reasoning features and its first official vision language model. The new 2B and 8B Instruct models let developers toggle reasoning via a simple API parameter to save compute, while Granite Vision 3.2 2B targets enterprise document understanding.
Wonder Dynamics just rolled out Motion Prediction, a new AI model designed to solve occlusion issues in markerless motion capture. It accurately predicts character poses even when actors are hidden behind objects or framed out, reducing manual cleanup for animators.
Alibaba's Tongyi Lab just released Wan2.1, an open-source video generation suite that beats top models like Sora on key benchmarks while running 2.5x faster. It supports bilingual text rendering and video-to-audio, and includes a lightweight 1.3B model that runs locally on an RTX 4090 GPU.
Hume AI just launched Octave, a speech-language model designed to understand emotional context instead of just reading words aloud. It detects sarcasm, panic, and character nuances straight from scripts or prompts, outperforming ElevenLabs in blind comparisons for audio quality and naturalness.
xAI just rolled out an uncensored voice mode for Grok 3 subscribers that goes where ChatGPT won't. Users can toggle personalities like an "unhinged" mode that curses and screams before hanging up, or an 18+ "Sexy" mode built for phone sex. Why it matters: While OpenAI keeps tight guardrails on voice AI, xAI is doing the exact opposite by letting its model run completely wild.
OpenAI just rolled out a free preview of ChatGPT's Advanced Voice mode to all users, powered by its cheaper GPT-4o mini model. It mimics the natural flow and tone of the flagship version, though audio time is capped by daily limits that alert you with 3 minutes left.
Apptronik just inked a pilot deal with manufacturing giant Jabil to deploy its Apollo humanoid robots on factory floors for sorting and transport tasks. If all goes smoothly, Jabil will eventually mass-produce Apollo in its own facilities—meaning the robots will literally help build themselves, with commercial manufacturing targeted for 2026.
Poe just launched Poe Apps, letting users create and share interactive visual applications simply by describing them. Powered by Claude 3.7 Sonnet, the tool writes the UI and JavaScript while letting creators chain together models like GPT-4o and Flux. The apps are shareable on the web, with mobile support and creator monetization planned next.
Microsoft just removed all usage limits on Copilot's Voice mode and OpenAI o1-powered Think Deeper reasoning tool for free users. Anyone can now hold uninterrupted voice conversations or work through complex logic problems without hitting query caps. Copilot Pro remains at $20 a month for peak-time priority.
Adobe just launched a full-fledged Photoshop app for the iPhone, leaving behind the watered-down Photoshop Express experience. The free tier packs real layers, masks, and Firefly-powered Generative Fill, while advanced tools like the Magic Wand require a subscription. Why it matters: Mobile creators finally get genuine desktop-style editing on the go.
OpenAI just published the system card for its agentic Deep research feature, breaking down the safety guardrails built into the tool before its launch to Pro users. Powered by an early version of OpenAI o3, the model scored a medium rating across four frontier risk areas, with safety teams focusing heavily on web-based prompt injections and data privacy.
Google just launched Gemini 2.0 Flash-Lite for general availability across the Gemini API, Google AI Studio, and Vertex AI. The lightweight model beats 1.5 Flash on reasoning and math benchmarks while offering cheaper long-context pricing at $0.10 per million input tokens.
DeepSeek is accelerating the rollout of R2, the successor to its hit R1 reasoning model, according to people familiar with the matter. The startup originally targeted an early May release but now wants it out as soon as possible, aiming to deliver better coding skills and broader multilingual reasoning. Why it matters: If DeepSeek strikes twice on a budget, it will put even more pressure on Western rivals racing to keep up.
Chinese tech giants including Tencent, Alibaba, and ByteDance are reportedly ramping up orders for Nvidia H20 chips to keep up with DeepSeek's low-cost models. While investors initially feared efficient models would gut hardware demand, heavy inference usage across sectors like healthcare and education is driving massive server orders instead.
A supergroup of 1,000 musicians just dropped a silent album called 'Is This What We Want?' to protest British copyright proposals. The planned rules would allow tech firms to scrape creators' work for AI training unless artists proactively opt out. The release features ambient room noise from empty studios, with 12 track titles spelling out a warning against legalizing data theft.
