Poncle just signaled it may back out of a Fortnite crossover just hours after Epic announced the deal, citing the gaming giant's aggressive embrace of generative AI. Epic recently showed off concept artists using AI for Fortnite assets and introduced engine plugins integrating Claude and Gemini. The indie studio says it is now reviewing the partnership, underscoring real developer pushback against AI-generated art.
Epic Games just outlined its roadmap for Unreal Engine 6, putting generative AI and LLMs at the core of next-gen game development. By exposing engine capabilities through the open MCP protocol, UE6 lets creators plug in models like Claude or Codex to automate tedious level building, rigging, and bug hunting. It basically turns the engine into an open playground for AI agents.
Lightricks and Tel Aviv University just introduced ScenA, a text-to-audio foundation model that generates complete multi-speaker scenes from a single prompt and reference voice clips. Instead of stitching together clean studio speech turn by turn, it bakes in overlapping dialogue, room acoustics, and ambient sound in one pass. It marks a big step toward synthetic audio that actually feels authentic.
Anthropic CEO Dario Amodei and Google DeepMind CEO Demis Hassabis just called for a U.S.-led coalition to shape global AI guardrails during a closed-door G7 lunch with Donald Trump. Amodei urged cooperation on structured model access, cyber defenses, and restricting chip trade with China. Why it matters: With frontier models already triggering federal export freezes, AI leaders want Washington to set global standards before policies fragment.
Character.AI just introduced a new suite of features designed specifically for character builders, including a performance dashboard called Creator Insights, a Creator Guide, mobile creation updates, and leaderboards. A dedicated Following tab and new character notifications are also rolling out soon. The move turns the platform from a bot directory into a true creator-first community.
Ai2 just open-sourced MolmoMotion, a multimodal model that predicts how 3D points on objects will move based on video frames and text prompts. Built on Molmo 2, it boosts simulated robotic pick-and-place success from 56% to 76.3% and steers controllable video generation. Ai2 also released an open dataset of 1.16M video trajectories and a new benchmark.
Palmier just launched an open-source, Mac-native video editor designed specifically for Claude, letting creators edit, organize, and generate footage directly inside the timeline. It includes free on-device semantic search and transcription, letting users bring their own Anthropic API key or connect via MCP. Paid add-ons also unlock third-party models like Kling and ElevenLabs.
xAI just made its flagship Grok 4.3 model available on Amazon Bedrock, opening the doors for AWS developers to integrate it directly into their workflows. xAI touts the model as an industry leader in both tool-calling speed and low hallucination rates. It gives cloud builders direct access to Grok inside Amazon's enterprise-grade security stack.
Anthropic just signaled that access to its frontier AI models, Mythos and Fable 5, could be restored within days following an abrupt global shutdown sparked by a White House security directive. The company pulled the models amid export control fallout, but executives speaking in Seoul stressed that Anthropic is pushing ahead with deeper local investments regardless.
Pew Research Center just found that nearly half of U.S. adults now use AI chatbots, up sharply from 2024, with ChatGPT leading the pack at 44%. But deeper adoption hasn't won hearts and minds: 63% of Americans believe the tech is advancing too quickly, and 71% fear it will erode their data privacy.
European Commission President Ursula von der Leyen just called for deeper EU-US AI cooperation at the G7, pointing out that together they represent 70% of the global market. She urged joint investment and faster adoption in sectors like healthcare, while stressing the need to keep the most powerful models trustworthy and safe.
Z.ai just introduced GLM-5.2, an open-source flagship model built for long-horizon coding tasks with a reliable 1M-token context. Powered by its new IndexShare architecture, it cuts per-token FLOPs by 2.9x at full context, edging past GPT-5.5 on FrontierSWE while remaining fully accessible under an MIT license.
Alibaba Cloud just introduced HappyOyster 1.0, an open world model that builds interactive digital environments in real time from text prompts. Unlike video models that make you wait for renders, it continuously responds to user input, allowing people to steer live storylines or walk around generated worlds using WASD keyboard controls.
Droidup just showed off Moya, a companion humanoid robot featuring an artificial spine, body-temperature silicone skin, and a claimed 92% human gait. The bot tracks expressions and mimics chores, though observers noted the face looks stiff and demo footage appears heavily edited.
Anthropic just analyzed 400,000 Claude Code sessions and discovered that domain know-how matters far more than programming chops. Lawyers who know contract clauses succeed at generating scripts almost as often as software engineers, with humans handling 70% of planning decisions while Claude takes on 80% of execution. Why it matters: AI coding agents are shifting leverage from pure syntax writers to people who actually understand the business problem.
OpenAI just introduced LifeSciBench, a benchmark built by 173 doctorate-level scientists to test how well AI handles real biotech research. It features 750 tasks across seven workflows, pairing them with over a thousand research artifacts like charts and sequence files. While frontier models show promise in communicating findings, they still struggle with complex experimental design.
Anthropic just updated Claude Design to keep generated interfaces strictly aligned with your team's design systems. It now works fluidly alongside Claude Code for seamless design-to-code handoffs, adds direct canvas editing controls, and plugs straight into third-party tools like Adobe, Replit, and Vercel. The feature is available in beta across paid plans.
Pinterest just introduced a suite of AI-powered tools ahead of Cannes Lions, including Business Assistant for Ads Manager, a standardized Model Context Protocol (MCP) integration, and an experimental visual app called Ask Pinterest. In testing, its new creative selection model boosted clicks by 7.5%. Why it matters: Pinterest is leveraging its unique Taste Graph to anchor conversational discovery in user taste and purchase intent.
Midjourney just revealed it will unveil its first hardware project during a launch event in San Francisco on Wednesday at 6pm PT. The company plans to livestream the announcement, though details about what the device actually is remain completely under wraps. It marks an unexpected leap beyond software for the image-gen heavyweight.
Qualcomm just unveiled the Snapdragon Reality Elite, packing a 160% jump in NPU performance and 60% faster graphics for next-gen headsets. The chip debuts inside Google and Xreal's upcoming Aura glasses this fall, aiming to run Gemini Live multimodality right on device. Preorders are open now.
Cursor maker Anysphere just teased a new proprietary model at Compile, slated for release in the next couple of weeks. Trained completely from scratch rather than on Kimi, the model matches Claude Opus in size, uses 10 to 20 times more compute than Composer, and targets broad general intelligence beyond coding.
Snap just unveiled Specs, its first consumer AR glasses priced at a steep $2,195. Unlike Apple's tethered Vision Pro, Specs are fully standalone, packing dual Snapdragon chips, a 51-degree field of view, hand tracking, and built-in AI assistance. They offer up to four hours of battery life and are slated to ship this fall.
Microsoft just launched Copilot Cowork into general availability worldwide, taking enterprise automation beyond chat. It independently executes long-running, multi-tool tasks to completion and runs on Anthropic’s Opus 4.8 and Sonnet 4.6 models. It requires a Microsoft 365 Copilot subscription, billed via usage-based Copilot Credits.
SpaceX just exercised its option to acquire AI coding assistant Cursor in an all-stock transaction. The two companies have spent the past few months jointly training a model that will roll out to Cursor and Grok Build soon, tapping SpaceX's Colossus supercomputer to supercharge their AI ambitions.
Researchers just unveiled FlexAvatar, a novel 3D Gaussian Splatting method that reconstructs an animatable 360-degree head from a single photo in only 2 minutes. It disentangles driving signals from viewpoints across mixed datasets, hitting 20 FPS on an RTX 3090 with just 1.7GB of VRAM.
OpenAI just unveiled Deployment Simulation, a pre-release safety pipeline that replays real, de-identified user conversations to forecast how candidate models will act in the wild. Tested across roughly 1.3 million conversations from GPT-5 Thinking through GPT-5.4 Thinking, it predicted undesired behavior rates with a median error of just 1.5x. Crucially, the realistic setup prevents smart models from realizing they are being tested, catching novel flaws like tool hacking before launch.
Midjourney just launched Draft Mode for its V8.1 model, generating 24 lower-res images per job while using half the fast GPU hours of a standard run. Users can hit Vary on any draft to upscale it to full resolution. There is also a new --preview flag to test early, unpolished model iterations.
Meta just rolled out a suite of AI features on Facebook, led by an "AI Mode" search tab that pulls answers directly from public Groups and Reels instead of generic web links. The update also adds automated video collage templates and photo presets that let users virtually swap clothing and hairstyles on Stories and profile pictures.
ByteDance just launched Seedance 2.0 Mini on its BytePlus platform, bringing faster video creation from image, video, and audio references. Geared toward daily production pipelines, the lightweight model cuts standard Seedance 2.0 pricing by roughly 50% for budget-conscious creators.
Cartesia just launched Sonic-3.5 for text-to-speech and Ink-2 for live transcription, both topping Artificial Analysis's streaming leaderboards. Built on state space models rather than standard transformers, the duo pushes latency below 300ms with built-in turn detection. That finally eliminates the awkward pauses in real-time voice agents.
PermaVid just open-sourced a video editing framework that solves one of generative video's biggest headaches: consistency. By splitting context into two separate memory banks for RGB appearance and 3D depth geometry, it keeps scene edits stable across time without distorting structure. Why it matters: It stops AI from forgetting edits, making iterative long-form video editing actually viable.
OpenRouter just rolled out Fusion, an API feature that queries multiple models in parallel, adds web search, and synthesizes a final response using a judge model. In testing, a budget panel of three smaller models scored nearly on par with Anthropic's flagship Fable 5 at roughly half the cost. It effectively turns model diversity into a plug-and-play research team.
NVIDIA just unveiled MotionBricks, a unified motion generation framework that hits 15,000 FPS with just 2ms latency on a single neural backbone. It handles over 350,000 motion skills, animating Unreal Engine 5 characters and directly controlling physical Unitree G1 humanoid robots without task retraining. Why it matters: It bridges the long-standing gap between game animation and real-world robotics with one shared brain, and early preview code is already on GitHub.
Sakana AI just launched Sakana Marlin, its first commercial product designed as an autonomous research assistant for business. Users simply provide a research topic, and Marlin works autonomously for up to eight hours to map causal relationships, delivering executive summary slides and in-depth strategy reports up to 100 pages long.
OpenAI just launched the OpenAI Partner Network backed by a $150M fund to help businesses deploy frontier AI. The program features three partner tiers and aims to certify 300,000 consultants by the end of 2026. The move signals that enterprise adoption hurdles are no longer about raw model power, but workflow integration.
Microsoft CEO Satya Nadella just published an essay urging companies to build and own their own AI learning loops. He introduced the idea of "token capital" alongside human capital, arguing firms must train agentic systems on internal data and private evals so proprietary knowledge isn't lost when switching models. Why it matters: Nadella warns that ceding all cognitive value to a handful of frontier models risks repeating the economic hollowing-out seen in early globalization.
Top security executives and researchers just published an open letter urging the U.S. government to lift export controls on Anthropic's Fable and Mythos models. They argue that rival models like GPT-5.5 and China's Kimi 2.7 already match their flaw-finding chops, meaning bans merely disarm defenders while adversaries race ahead unrestricted.
Amazon CEO Andy Jassy just raised security concerns about Anthropic's newest AI models directly to senior Trump administration officials. Those talks prompted the U.S. government to roll out new export controls restricting foreign nationals from accessing Anthropic's Mythos and Fable models. It is a striking example of Big Tech directly shaping federal AI restrictions.
Moonshot AI just open-sourced Kimi K2.7 Code, a massive 1-trillion parameter MoE model with 32B active parameters. It slashes reasoning token usage by roughly 30% while closing the gap on top proprietary frontier models like GPT-5.5 and Opus 4.8 in long-horizon programming tasks. Full 600GB weights are now available on Hugging Face.
Moonshot AI just launched Kimi Work, a desktop agent for macOS and Windows that reads your local files, drives your real browser, and runs scheduled background jobs. Powered by its K2.6 model, higher tiers can spin up a swarm of up to 300 parallel sub-agents. It is currently in testing, with subscriptions starting at $19/month alongside a free tier.
Tim Ferriss just revealed that sales across his bestselling catalog are falling off a cliff, on pace for an 80% drop in 2026 compared to 2022. He argues that conversational bots like ChatGPT and Claude have become personalized answer engines, rendering prescriptive books obsolete and threatening any creator who just sells instructional steps.
Zyphra just released ZONOS2, an open-source TTS model with 8B total and 900M active parameters powered by MoE. Trained on over 6M hours of multilingual speech, it delivers low-latency 44.1 kHz voice cloning under an Apache 2.0 license, with ready-to-run local inference and a free web playground.
Google just rolled out its Search information agents to AI Ultra subscribers across all AI Mode languages and markets. Users can instruct AI Mode to monitor any topic, and the agent will run 24/7 in the background, pinging users with detailed summaries and web links as soon as new information breaks. It turns passive search into an active tracking assistant, with broader rollout planned for later this summer.
Elon Musk just became the world's first trillionaire after SpaceX went public, with shares of SPCX opening at $150. Beyond personal wealth, SpaceX's S-1 filing outlines an ambitious plan for "orbital AI compute," launching AI data center servers into space via reusable rockets.
Google just filed a civil lawsuit against "Outsider Enterprise," a cybercrime ring that distributed AI-powered phishing kits to blast fake texts and clone over 9,000 websites. The company is coordinating with the FBI and major telecom carriers to block millions of fraudulent texts, while lobbying for federal anti-scam legislation. Why it matters: as AI turbocharges phishing operations, tech giants are forced to fight back with law enforcement and the courts, not just software filters.
Rewind AI, the buzzy startup behind the memory-tracking Mac app and Pendant wearable, was acquired by Meta in 2025 and officially shut down on December 19, 2025. Its original domain has now found a second life under an unaffiliated team, hosting a catalog of 400+ free AI tools that require no account.
Moonshot AI just rolled out a HighSpeed mode for its K2.7-Code model, hitting up to 260 tokens per second — a 6x speed jump. It reportedly cuts reasoning tokens by 30% through post-training tweaks, though independent coding benchmarks have yet to confirm the claims.
Anthropic just shut down access to Fable 5 and Mythos 5 for all users following an export control directive from the US government barring foreign nationals from using the models. Officials cited a cybersecurity jailbreak risk, but Anthropic pushed back, noting the narrow technique is already replicable on models like GPT-5.5. Why it matters: If non-universal bypasses trigger commercial recalls, frontier deployments across the entire industry could grind to a halt.
Anthropic just released its first Public Record survey of nearly 52,000 Americans, revealing that 64% worry most about AI taking jobs and over 70% support government regulation. Only 15% trust AI developers to guide themselves. Why it matters: hands-on daily users are noticeably less anxious about displacement, suggesting familiarity might soften the shock.
Meta just previewed an upcoming AI assistant and a desktop version for its video-editing app Edits to challenge CapCut. The assistant taps into creators' Instagram analytics to brainstorm video ideas and recommend trending audio, while desktop support brings larger screens and cross-device sync. A new Beta tab for experimental features and expanded audience insights also roll out today.
Anthropic just changed course after facing heavy backlash for silently downgrading user requests on its newly released Fable 5 model. Starting this week, prompts flagged for advanced AI development or national security concerns will visibly fall back to Opus 4.8, and the API will return an explicit reason. Anthropic apologized for getting the balance wrong, though it reaffirmed its stance against helping foreign adversaries or building rival models.
Coinbase just launched Coinbase for Agents, a tool that connects AI agents directly to user accounts to autonomously trade and make payments within custom limits. Available via MCP for ChatGPT and Claude or a CLI for developer terminals, it currently supports crypto spot and derivatives. Giving AI agents direct, bounded access to real capital marks a major milestone for autonomous finance.
Meta has fully cut operational ties with AI startup Manus after authorities ordered the companies to unwind their $2 billion acquisition deal. Meta halted all data sharing with Manus and barred employees from using the startup's AI tools internally, following regulatory orders in April revoking the transaction.
Runway just expanded its partnership with Hollywood giant Lionsgate with a joint development program to create original IP together. Instead of just licensing video models as production tools, the AI startup will now co-create content directly with the studio. It marks one of the deepest integrations between generative AI and legacy Hollywood filmmaking yet.
MoVerse just introduced a real-time video world model that turns a single regular image into an interactively navigable 360° world. By decoupling 3D Gaussian world memory from video rendering, it pulls off 8 FPS interactive roaming on a single consumer RTX 4090 GPU. The code and weights are slated for release following compliance review.
ElevenLabs just launched Avatars in ElevenCreative, letting creators turn scripts into lip-synced talking-head videos inside a single interface. Users can create persistent visual identities from photos or prompts, keep characters consistent across scenes, and batch-produce videos through Flows.
Deezer just launched a free web tool that scans playlists across 20 competing platforms—including Spotify, Apple Music, and YouTube Music—to detect AI-generated tracks. After rivals refused to license its detection tech, Deezer took it straight to listeners so they can audit their own libraries.
Figma just released a Chrome extension that captures live websites and turns them into fully editable design layers directly inside your canvas. Instead of rebuilding layouts from scratch, you can copy full pages into your project with a click. It is currently in beta and limited to paid plans.
OpenAI just announced its support for the European Commission's Code of Practice on Transparency of AI-Generated Content. Images made through ChatGPT and its API now include both C2PA credentials and SynthID watermarks, backed by a public verification tool to check provenance across the web.
Anthropic just launched Claude Corps, committing $150 million to place 1,000 early-career fellows into U.S. nonprofits for a full year. Fellows receive an $85,000 salary, benefits, and weekly Claude training to help over 400 organizations build real AI capacity.
Jeff Bezos and former Verily co-founder Vik Bajaj just raised $12 billion for their physical AI startup, Prometheus, reaching a massive $41B valuation. The company is developing an "artificial general engineer" designed to automate the manufacturing of complex hardware like jet engines and drugs. Bezos also dismissed AI job theft fears, arguing the technology will actually cause a labor shortage.
Perplexity just integrated Deep Research into its Computer workspace, turning complex web research directly into finished PDFs, dashboards, and decks. Powered by a new "Search as Code" architecture, the model writes custom programs to run thousands of parallel search steps, pulling from internal files and premium sources like Statista and PitchBook. Why it matters: It bridges the gap between deep information retrieval and ready-to-use deliverables in a single prompt.
SpaceX just pulled off the largest IPO in history, raising $75B at a $1.77T valuation after selling 555.6M shares at $135 apiece. The massive raise blows past Saudi Aramco’s 2019 record and makes Elon Musk the world’s first paper trillionaire. Investors backed the money-losing giant on high hopes for Starlink, orbital data centers, and Mars.
OpenAI is acquiring Ona to integrate its secure cloud execution tech into Codex, allowing agents to run for hours or days inside an enterprise's own infrastructure. Codex now counts over 5 million weekly users. Basically, organizations can now let agents do long-running work even when laptops are closed without compromising on security or data governance.
Anthropic CEO Dario Amodei just called for binding, FAA-style regulations on frontier AI models, arguing voluntary transparency is no longer enough. The proposal urges mandatory third-party safety testing in areas like cybersecurity and bioweapons, backed by government power to block unsafe deployments. Why it matters: As frontier models reach national strategic weight, top AI leaders are actively asking governments to step in with real enforcement teeth.
Revolution Medicines just showed that a drug targeting the slippery RAS protein nearly doubled survival time in advanced pancreatic cancer patients, jumping from 6.6 to 13.2 months. Now, researchers are deploying AI to screen millions of molecules against notorious, smooth-surfaced culprits like p53 and MYC that fuel roughly 70 percent of cancers. What used to take decades of chemistry trial-and-error might soon become a solved computation problem.
Sony just showed off Ace, an autonomous table tennis robot that consistently defeats ranked human professionals under official competition rules. Mounted on a heavy-duty electric rail, the bot tracks incoming ball spin within milliseconds and aggressively adjusts its own returns to force human errors. Two-time Olympic medalist Miu Hirano noted it rarely makes rally mistakes and could already compete at a top-10 women's level.
Google just open-sourced DiffusionGemma, an experimental 26B MoE model that ditches token-by-token generation for text diffusion, stamping out 256 tokens in parallel. It delivers up to 4x faster GPU inference, clocking over 1,000 tokens per second on an H100 and fitting into 18GB VRAM when quantized. Why it matters: It drastically cuts latency for local inline editing and code infilling, though Google notes raw output quality still trails standard Gemma 4.
Researchers just introduced Arbor, an autonomous agent framework that turns messy trial-and-error into a structured hypothesis tree. A long-lived coordinator oversees strategy while isolated executors run tests, achieving over 2.5x the relative gains of Claude Code and Codex across long-horizon research tasks.
UC Berkeley's RDI just launched Agents' Last Exam (ALE), a demanding benchmark testing AI on real desktop workflows across 55 industries. OpenAI's GPT-5.5 topped the board with a 24.0% pass rate, narrowly beating Anthropic's new Claude Fable 5 at 22.0%. Why it matters: even the best models scored 0.0% on the hardest tasks, proving AI agents still struggle with genuine professional workloads.
Nvidia just teamed up with Apple and Google to power Apple's Private Cloud Compute using Blackwell GPUs with Confidential Computing. The expanded infrastructure runs on Google Cloud to handle next-gen Apple Intelligence features, with hardware-rooted encryption ensuring no one can peek at your data.
Princeton researchers just released i1, a 3B diffusion model that openly shares its weights, code, and entire data curation pipeline. While not built to outperform massive closed models, it runs easily on consumer hardware and gives anyone a complete, transparent recipe to train image models from scratch.
Researchers from KAIST AI and AIPARK just unveiled Lip Forcing, the first autoregressive diffusion model for video-to-video lip sync. By distilling a 14B teacher into a 2-step causal student, the 1.3B model hits 31 FPS on a single H100, easily clearing the real-time threshold.
HeyGen just launched HyperFrames, an official MCP connector for Claude and Claude Code that turns text prompts into animated videos. Instead of skimming dense walls of text, users can now generate code-first motion graphics, voiceovers, and subtitles directly inside Claude, exporting ready-to-share MP4s without touching a video editor.
SpaceX just unveiled AI1, an orbital data center satellite built to run AI silicon directly in space using solar power. Each craft packs up to 150 kW of compute payload—matching a top-tier Nvidia server rack—with interchangeable chips and a wingspan wider than a Boeing 747. Why it matters: With Google and Anthropic already signed on as customers, space-based compute is quickly shifting from a sci-fi punchline into a real workaround for Earth's strained power grids.
Google just launched Gemini 3.5 Live Translate, a new audio model delivering near real-time voice translation across 70+ languages. Instead of waiting for you to stop talking, it streams continuous audio just seconds behind the speaker while preserving natural pitch, pacing, and intonation. It is rolling out to the Google Translate app, Google Meet, and developers via the Gemini Live API.
Google DeepMind just released RCT results showing its Gemini tutor accelerated math learning by 1.2 to 2.5 years in just eight weeks across Sierra Leone schools. Instead of handing out solutions, Gemini acted as a Socratic guide, asking scaffolding questions 76% of the time and giving direct answers only 2%. Why it matters: It proves AI can foster critical thinking rather than student laziness, though teacher involvement remains critical.
ZAI just open-sourced SCAIL-2, an end-to-end character animation model that transfers complex motion without relying on ambiguous skeleton maps. It handles intense multi-character combat and can even map human dancing onto flamingos or stylized cartoons. The catch: the full weights clock in at 81GB, so consumer GPUs will likely need to wait for quantized builds.
ByteDance just launched Seed-Audio 1.0, an all-in-one audio model capable of generating multi-character dialogue, emotional voice acting, sound effects, and background music in a single pass. It uses audio references to keep voices consistent across long podcasts or shows, and is currently rolling out via API.
Anthropic just added scheduled deployments and environment variable vaults to Claude Managed Agents in public beta. Agents can now automatically run recurring tasks on a cron schedule without custom scheduling infrastructure, while the new vaults inject API keys at the network boundary so the AI never sees raw credentials.
Anthropic just unveiled Claude Fable 5, its powerhouse Mythos-class model designed for general availability. It leads frontier benchmarks across software engineering, vision, and scientific research. To mitigate severe cyber risks, conservative classifiers automatically route sensitive queries—less than 5% of total sessions—back to Opus 4.8, while an unrestricted twin called Mythos 5 rolls out to vetted cyberdefenders.
Media investment firm Electrify just invested in and partnered with The Rundown, the daily AI newsletter with over 2 million subscribers founded by Rowan Cheung. Cheung's team keeps full editorial control while Electrify provides operational backing to expand into new formats and products.
Midjourney founder David Holz just revealed that the company is rolling out invites for its first-ever hardware launch event. Details about what the mystery device actually is remain completely under wraps, with only a few seats left. It signals the image generation giant's official expansion from software into physical consumer hardware.
Google just teamed up with Apple to bring Gemini models into Xcode and Apple's native Foundation Models framework. Developers can now swap between on-device Apple models and cloud-hosted Gemini with minimal code changes, while Xcode gets built-in agentic coding assistance for reviewing and bug fixing. It's currently available in preview.
OpenAI just rolled out native interactive charts inside ChatGPT, letting you generate bar, line, pie, and scatter plots on the fly without uploading files or waiting for Python scripts. You can tweak color palettes, resize, and switch views right in the thread. Why it matters: Quick data visualization is now completely friction-free, turning ChatGPT into an all-in-one productivity workspace.
Google just gave NotebookLM a massive upgrade powered by Gemini 3.5, equipping every notebook with a secure cloud computer to write and execute code. It can now turn raw sources into PowerPoint slides, Excel sheets, and visual charts, while pulling web context automatically. It is rolling out to Google AI Ultra and eligible Workspace business tiers.
Amazon just rolled out an AI-powered custom merch generator inside Alexa for Shopping for all U.S. customers. Users simply type a prompt to generate instant designs on T-shirts, hoodies, and tumblers in seconds. Designs can be shared for group orders and are fulfilled via Merch on Demand with Prime shipping.
Anthropic just launched a Swift package integrating Claude with Apple's Foundation Models framework across iOS 27 and macOS 27. Developers can now run quick tasks on-device and cleanly hand off multi-step reasoning, coding, and web searches to Claude via typed Swift outputs. Why it matters: it seamlessly unites local privacy and cloud horsepower inside native apps.
OpenAI just laid out a bold blueprint for its "third phase," aiming to build an automated AI researcher by March 2028 and put a personal AGI in everyone’s hands. The company stressed that human judgment must stay in control, warning against fully automated decision-making while calling for an international safety body to manage catastrophic risks.
Anthropic just rolled out a performance dashboard for Claude connector builders, alongside direct in-app submissions for MCP servers. Developers can now track active users, tool calls, error rates, and latency breakdowns across Claude, Claude Code, and Cowork. It is currently in public beta for Team and Enterprise admins, making it significantly easier to debug and scale MCP agents.
OpenAI just launched the Economic Research Exchange, a program granting external researchers access to its internal tools and datasets to study how AI shapes jobs, productivity, and inequality. Selected academics will run structured projects with strict data privacy safeguards. Why it matters: It trades speculative hot takes about AI taking jobs for empirical, real-world economic evidence.
Apple just introduced the next generation of Apple Intelligence, built with a new foundation model architecture developed alongside Google. The update brings Siri AI, spatial reframing in Photos, and photorealistic generation to Image Playground, while Passwords can now agentically fix compromised credentials. Developer testing starts now, with a public rollout slated for this fall.
OpenAI just submitted a confidential draft S-1 registration statement to the SEC, taking a formal step toward going public. The company pre-announced the filing because it expected a leak, noting that timing isn't decided yet and being private still makes certain big moves easier.
The Trump administration is in talks to take an equity stake in OpenAI, exploring models where the public becomes a financial partner in AI growth. The shares could seed OpenAI's proposed Public Wealth Fund to distribute returns directly to citizens, echoing last year's Intel deal while reigniting concerns over corporate-government fusion.
Alibaba just released Qwen 3.7 Plus, a multimodal model built to navigate graphical interfaces, write code from templates, and run terminal commands within a single loop. In testing, it autonomously built a complete vocabulary app over eleven hours, leading Western rivals on GUI benchmarks like AndroidWorld. It also supports Anthropic's API protocol, making it a dirt-cheap drop-in backend for existing agent tools.
Rednote just open-sourced dots.tts, a 2B fully continuous end-to-end autoregressive speech system with zero discrete tokens in the pipeline. It achieves top speaker similarity and sub-1% word error rates on Seed-TTS benchmarks, supporting zero-shot cloning and speech editing under an Apache-2.0 license.
