BFL just released FLUX.1 Tools, adding inpainting, outpainting, depth guidance, canny edges, and variation generation to its text-to-image models. The models are available as open-access weights and API endpoints, though Depth and Canny were later deprecated. It gives creators far more precision over generated scenes.
Perplexity just launched an Election Information Hub to help voters navigate key issues, polling logistics, and live race results. The platform taps into data from The Associated Press and Democracy Works' API, offering source-backed summaries on candidates and ballot measures. It is a bold test of whether AI search can handle high-stakes civic info without hallucinating.
Amazon just delayed its generative AI-powered Alexa revamp into 2025, quietly shuttering its beta program in the process. Testers reported the new assistant was slow, hallucinated, and struggled with basic smart home commands like setting kitchen timers or turning off lights. Why it matters: Building a conversational bot is one thing, but reliably connecting an LLM to legacy smart home hardware is proving to be a massive headache.
Perplexity is quietly testing an in-platform shopping feature called "Pro Shop" that lets users research and buy products directly without leaving the site. Eligible items feature a "Buy with Pro" button with promised free shipping, turning the AI search engine into a full checkout loop.
ElevenLabs just rolled out an open-source project called X to Voice alongside a new Voice Design API. The tool scans any X profile's bio and posts to craft a prompt, generating three unique voice previews and even animating profile pictures with matching lip movements. Developers can now also build custom voices for their AI characters via the API.
Anthropic just launched official Mac and Windows desktop apps for Claude in public beta, alongside a new voice dictation feature on iOS, Android, and iPadOS. The dictation tool lets users record up to 10 minutes of audio that Claude then transcribes and answers. It isn't a live, conversational voice mode like ChatGPT's, but it makes firing off complex prompts on the go a whole lot easier.
Anthropic just launched official Mac and Windows desktop apps for Claude in public beta, available to both free and paid users. The dedicated app lets users invoke the assistant via keyboard shortcuts for focused work. Alongside the desktop rollout, mobile versions also gained a native voice dictation feature supporting recordings up to 10 minutes.
Google just rolled out generative AI capabilities for Google Maps Platform and Google Earth, allowing developers to ground LLMs in real-world data from 250 million places. Google Earth is also getting Gemini-powered multi-step reasoning to slash urban planning analysis time from days to minutes. Next month, Rivian will begin rolling out AI place summaries to its vehicle screens.
OpenAI just built web search directly into ChatGPT, serving up real-time answers with clickable source citations. Instead of sending you off to a traditional search engine, it blends conversational chat with live sports scores, weather, and breaking news. Why it matters: It turns ChatGPT into a full-blown search engine alternative backed by official publisher partnerships.
OpenAI just rolled out its Advanced Voice Mode to ChatGPT desktop apps on macOS and Windows, bringing human-like, low-latency audio conversations directly to your computer. Users can chat back and forth while coding or brainstorming over uploaded files with five new voices. It is currently limited to paid Plus and Team subscribers, excluding the EU and UK.
Wonder Dynamics just rolled out a new tool that converts multi-camera live-action video straight into fully editable 3D scenes. Instead of generating flat video, it outputs industry-standard 3D assets that creators can light, tweak, and reframe inside Blender, Maya, or Unreal. Why it matters: Filmmakers can now shoot rough takes in a living room and instantly produce clean previsualizations.
Researchers just revealed that OpenAI’s Whisper transcription model frequently hallucinates, with a University of Michigan study finding fabricated text in 80% of reviewed transcripts. Other engineers found widespread inventions, including fictional medical treatments. Why it matters: Whisper is already widely deployed across medical consultations and enterprise cloud platforms.
Apple is quietly prepping an AI-powered system that generates review summaries for apps on the App Store, highlighting the most common customer feedback before you hit download. The summaries update as new reviews roll in and run server-side rather than on-device. Why it matters: You won't have to scroll through dozens of rants to see if an app actually works, though developers will be watching the accuracy closely.
OpenAI just rolled out chat history search for ChatGPT web users. Instead of endlessly scrolling down the sidebar to dig up an old prompt, you can now search directly through past conversations to find what you need.
Microsoft just pushed back against viral rumors claiming its controversial Recall AI feature is secretly being installed on Windows 11 24H2. The company reiterated that Recall remains strictly opt-in and fully uninstallable, with experts noting that recent system references are actually stubs meant to make removal easier for admins. Why it matters: After months of privacy pushback, users are on edge, and Microsoft is still working to rebuild trust before its delayed preview rollout.
Apple just previewed upcoming AI-powered features for Final Cut Pro during its M4 Mac mini launch. The update introduces a "Transcribe to Captions" tool that generates subtitles for a 10-minute video almost instantly, along with faster AI video effects. Apple has not yet announced a release date.
LinkedIn just launched Hiring Assistant, its first AI agent built to tackle recruiter grunt work. The tool converts rough notes into full job postings and sources candidates based on skills rather than degrees or location. It is currently available to select enterprise clients like AMD and Canva ahead of a wider rollout.
GitHub Copilot just added Anthropic's upgraded Claude 3.5 Sonnet as a model option directly within VS Code and GitHub.com. Powered via Amazon Bedrock, the top-tier coding model is rolling out in public preview to all Copilot Chat users and organizations over the coming weeks.
Google just announced its biggest AI Overviews rollout yet, expanding search summaries to more than 100 countries and territories. The move pushes the feature past 1 billion monthly users worldwide, alongside broader multi-language support for English, Japanese, Spanish, and more.
Apple just rolled out its first set of Apple Intelligence features through free updates to iOS 18.1, iPadOS 18.1, and macOS Sequoia 15.1. Users get systemwide writing tools for rewriting and summarizing text, a redesigned Siri that accepts text prompts, and a Clean Up tool in Photos to erase unwanted objects. Why it matters: it marks Apple's first official leap into consumer generative AI, though it's currently limited to US English.
Coinbase just launched Based Agent, a template that lets users build their own AI agent equipped with a crypto wallet in under three minutes. Built alongside OpenAI and Replit, the agent can autonomously execute trades, token swaps, and staking via smart contracts. Basically, Coinbase wants to make giving an AI bot its own bank account effortless.
Meta is rolling out a firmware and app update for its Ray-Ban smart glasses, bringing hands-free visual AI features announced at Connect. Users can now simply ask about objects in front of them to trigger photos, scan QR codes, dial phone numbers spotted by the camera, and set parking reminders. Live translation and video AI aren't live yet, but the glasses are taking another step toward always-on visual assistance.
Google just dropped a major AI update for Search and Lens. Users can now record a short video while speaking a question aloud, letting Gemini analyze the moving scene to generate an AI Overview. The update also brings voice queries for photos, song identification to Circle to Search, and AI-organized search pages. Video search is currently rolling out in English for Search Labs testers.
Captions just rolled out an AI-powered social media manager that scans business websites and automatically generates video marketing campaigns. The tool extracts site content and trends to build a posting schedule, then produces ready-to-publish Reels and TikTok clips using digital avatars. It is available now for Max and Scale subscribers starting at $25 a month.
OpenAI just launched Canvas, a dedicated side-by-side workspace for writing and coding projects. Instead of getting stuck in long chat threads, users can highlight specific sections for inline feedback, restore previous versions, and use quick shortcuts to debug code or adjust reading levels. It marks the first major visual interface overhaul for ChatGPT.
Amazon just announced a refreshed Fire HD 8 tablet featuring new generative AI tools like keyboard writing assistance, Silk browser webpage summaries, and prompt-based wallpaper generation. You don't have to buy the new hardware to try them either, as Amazon is rolling out these features to multiple older Fire HD and Max tablets later this month.
Rabbit just rolled out LAM playground, a cross-platform agent system for its R1 pocket companion. Users can now teach the multi-modal agent to navigate websites via the Rabbithole cloud portal, performing tasks like searching Google and adding items to Amazon carts. It still takes around 45 seconds and trips over CAPTCHAs, but it finally delivers on Rabbit's original core pitch.
Microsoft just rolled out generative fill and erase tools to Paint on Copilot Plus PCs, letting users add or remove objects with a quick brush stroke. The Photos app is also adding an on-device super-resolution tool that can upscale blurry pictures up to 8x in seconds. Quick image touch-ups just got a lot harder to justify paying Adobe for.
Bing just expanded its generative search experience to handle informational and complex queries. Instead of a typical list of links, Bing aggregates answers into a cohesive, structured layout with citations. The feature is currently rolling out in beta for US users.
Microsoft just introduced a fresh suite of on-device AI experiences for Copilot+ PCs and Windows 11. Highlights include Click to Do, an interactive overlay suggesting quick actions across any screen, alongside offline natural language file search and 8x photo upscaling. Recall also returns with opt-in defaults and biometric locks, rolling out to Windows Insiders this fall.
Adobe just released Photoshop Elements 2025 and Premiere Elements 2025, adding AI object removal, artificial depth of field, and simplified video color LUTs. Geared toward casual creators who hate subscriptions, both consumer apps are sold as a one-time $100 purchase (or $150 bundle) with full Apple M3 support.
OpenAI just launched an integrated model distillation workflow directly in its platform, combining output storage, evals, and fine-tuning into one place. Developers can now capture completions from advanced frontier models and use them to train cost-efficient smaller models without juggling messy third-party pipelines.
Microsoft just gave Copilot its biggest overhaul yet, introducing a natural Voice mode, a Vision feature that sees what you're browsing, and Think Deeper for complex multi-step reasoning. It also launched a morning audio briefing powered by licensed publishers like Reuters and FT. Vision and deeper reasoning are rolling out first via Copilot Labs to select users.
Wondershare just launched Filmora 14, packing over ten new AI-driven tools aimed at speeding up workflows. The standout Smart Short Clips feature auto-extracts vertical highlights with subtitles from long-form footage, alongside AI video enhancement, lip-sync translation, and multi-camera editing for Windows and Mac.
Google just shipped a streamlined mobile experience for Google AI Studio, stripping away clutter and putting models front and center. The redesign makes prompting and testing models significantly easier and faster while on the go.
Raspberry Pi and Sony just launched the Raspberry Pi AI Camera, a $70 module bringing edge vision processing to the popular single-board computers. Powered by Sony’s IMX500 sensor, it runs neural networks directly on the chip without needing extra GPUs or accelerators. Why it matters: It gives makers and developers a dirt-cheap way to run real-time computer vision locally.
Mileseey Golf just launched the GenePro S1, an AI-powered golf rangefinder that factors in weather conditions to compensate distances. Beyond standard elevation, it analyzes humidity, wind, temperature, and air pressure, while offering centimeter-level green readings. At $799.99, it basically packs a PGA-level caddie's playbook into your golf bag.
ChatGPT's new Advanced Voice Mode was strictly instructed never to sing or hum, but a software architect just bypassed those guardrails by turning music into a chord-guessing game. The AI joined in for a duet of The Beatles' "Eleanor Rigby," proving it knows melodies far better than OpenAI wants to admit amid copyright concerns.
Microsoft just overhauled Windows Recall's security architecture after months of privacy backlash. The AI tool is now strictly opt-in and fully uninstallable, isolating snapshots inside secure enclaves that require Windows Hello biometrics to decrypt on the fly. Why it matters: Microsoft is trying to salvage its flagship AI feature by turning it into a showcase for client-side security.
Microsoft just launched general availability for fine-tuning GPT-4o and GPT-4o mini on Azure OpenAI Service. The update also brings Meta's new Llama 3.2 vision models, the Phi-3.5 family, and a new Correction feature to catch hallucinations in real time before users see them. It makes tailor-made enterprise LLMs much easier and safer to deploy.
Google just rolled out Gemini-powered contextual Smart Replies to Gmail on Android and iOS. Instead of generic one-liners, the AI analyzes the entire email thread to draft detailed responses complete with greetings and sign-offs. Why it matters: It turns quick replies into genuinely usable drafts, though it is currently limited to English and paid Gemini subscribers.
Google just added support for public YouTube URLs and audio files directly into NotebookLM. You can now analyze lecture transcripts, connect insights across recordings, and easily share generated Audio Overviews via a public link.
Airtable just rolled out new enterprise AI capabilities to evolve beyond a spreadsheet tool into a full-scale AI platform. The update introduces App Library for standardized AI applications and HyperDB to handle over 100M records, moving past simple chatbots into parallel workflow automation.
Microsoft just introduced a "correction" preview in Azure AI Studio that flags inaccuracies and rewrites flawed outputs before users ever see them. It works by grounding generated text against source documents, though Microsoft admits using models to police other models isn't foolproof.
Figma just relaunched its generative AI app builder as First Draft after pulling it in June for cloning Apple's weather app. The tool now lets designers pick from four component libraries—ranging from low-fi wireframes to high-fidelity patterns—powered by off-the-shelf models like GPT-4 and Amazon Titan. It is currently rolling out in limited beta.
Spotify just expanded its generative AI Playlist beta to Premium subscribers in the US, Canada, and beyond. Users can type in natural language prompts to spin up a custom 30-song playlist and tweak the vibe with follow-up instructions. Why it matters: Music curation is shifting from fixed genre tags to conversational, mood-based requests.
Snap just expanded its partnership with Google Cloud to power Snapchat's My AI chatbot with multimodal Gemini models. Users can now send photos and videos to get instant help like translating street signs or picking healthy menu items, which Google says has already driven a 2.5x jump in U.S. engagement.
Duolingo just introduced Video Call and Adventures at its Duocon 2024 conference to make language practice more interactive. Duolingo Max subscribers can hold real-time, adaptive conversations with Lily across English, Spanish, and French, while new mini-games let learners practice everyday tasks like ordering coffee in a low-pressure game environment.
