|
You can generate and edit Google Docs files directly in Gemini chats now
Google Gemini users can now generate, preview, and edit a wide range of file formats—including PDFs, Google Docs, Google Sheets, slide decks, Markdown files, TXT files, and code—directly within their Gemini chats. PDF preview and editing is available on both web and mobile, while other file formats are currently limited to the web. The feature is rolling out to all Gemini users starting today.
|
|
Mistral releases weights for its trillion-parameter AI model, targeting cybersecurity
Mistral has open-sourced the weights of a new trillion-parameter AI model specifically designed for cybersecurity use cases. The company says the open model gives organizations a level of control over their cyber defenses that closed, proprietary models cannot provide. The release marks a significant step in applying large-scale open models to enterprise security.
|
|
Ex-Palantir execs launch Dream with $17M from a16z to build AI world models for construction
Former Palantir executives have launched Dream, a startup that has raised $17 million from Andreessen Horowitz to build AI world models tailored to the construction industry. Like large language models trained on text, Dream's model is trained on over 100 million square feet of building data sourced from US permit archives. The system is grounded in real physics and construction standards to provide practical, reliable outputs.
|
|
Meta quietly launches an AI fashion app called Style Sketch
Meta has quietly released Style Sketch, an AI-powered fashion app that connects to Instagram to build a virtual closet from photos users have already shared. Users can also manually add clothing items by snapping a photo, recording a video, or uploading images from their camera roll. The low-key launch suggests Meta is experimenting with AI-driven consumer lifestyle applications beyond its core social platforms.
|
|
Anthropic's IPO filing and AGI ambitions explained
Anthropic has filed its S-1 to go public on the Nasdaq, and the document reveals a business plan deeply tied to building AGI—what some are calling 'building God or die trying.' The filing appears to justify massive compute spending, with the company asking Wall Street to bet on a future where recursive AI development leads to transformative returns. Analysts at SemiAnalysis have been digging into the numbers to understand the story behind the figures.
|
|
Microsoft's AI future has fewer jobs
Microsoft CEO Satya Nadella's repeated signals about AI reshaping the workforce are materializing, as the company is now offering voluntary buyouts to around 60,000 employees across the US, UK, Canada, and other regions. The move is seen as a clear indication that AI is set to fundamentally reduce the size of Microsoft's workforce. Changes even extend to classic products like Notepad, which is being updated with AI features.
|
|
AWS will be a $1T business, Jassy says. But first, it will spend big.
Amazon CEO Andy Jassy has stated his ambition for AWS to become the first $1 trillion revenue business, positioning it as the dominant force in AI cloud computing. Achieving that goal will require continued heavy capital expenditure investment, Jassy said. The announcement underscores the escalating spending race among major cloud providers to capture AI workloads.
|
|
OpenAI-backed Thrive Holdings acquires ADT
OpenAI-backed Thrive Holdings is acquiring home security company ADT in a $15.3 billion deal, with plans to integrate AI into the home security business. The acquisition signals a major push to bring AI capabilities into consumer safety and smart home infrastructure. The deal represents one of the larger AI-driven acquisitions in the consumer sector to date.
|
|
US investigated Chinese spy plot aimed at Nvidia CEO
US authorities investigated a Chinese espionage plot targeting Nvidia CEO Jensen Huang, according to sources cited by the New York Times. A top executive of a Singapore-based AI cloud company accused of illegally selling Nvidia chips to China was also suspected of being a Chinese spy. The case highlights growing national security concerns around AI hardware and chip exports.
|
|
As search traffic dries up, Reddit announces it's shifting to a logged-in experience
Reddit is rethinking its open-web model and shifting toward a logged-in user experience as AI-powered search tools reduce the flow of organic search traffic to the platform. The move aims to protect Reddit's audience and business as AI transforms how people discover information online. The shift marks a significant strategic pivot for a platform that has historically thrived on open, indexable content.
|
|
You don't need to believe in a god to get a religious exemption from using AI at work
A University of Kentucky administrator named David Durman successfully obtained a religious exemption from being required to use AI tools in the workplace, setting a notable precedent. The case illustrates that religious exemption protections can extend to AI usage mandates, even without traditional theistic belief. The ruling raises broader questions about employee rights as AI adoption becomes increasingly mandatory in professional settings.
|
|
AI has come for Arizona's water
The rapid expansion of AI data centers and chip manufacturing facilities in Arizona is putting significant strain on the state's already scarce water resources. Facilities like TSMC's chip plant require enormous amounts of water to operate, raising concerns about long-term sustainability in a desert region already facing water scarcity. The story explores the environmental costs of the AI boom playing out in the American Southwest.
|
|
Wabi, the AI agent that builds mini apps, now valued at $1.25B after latest round
AI app Wabi, which describes itself as a 'personal agent' capable of building mini apps, has raised $55 million in its latest funding round, pushing its valuation to $1.25 billion. The startup is positioning itself at the intersection of AI agents and lightweight app creation. The unicorn milestone reflects continued investor appetite for AI-native productivity tools.
|
|
Deepgram acquires Pipecat-maker Daily to bolster its voice AI infrastructure
Voice AI company Deepgram has acquired Daily, the company behind the open-source real-time voice and video framework Pipecat. The acquisition is expected to give Deepgram deeper capabilities in real-time voice and video infrastructure, strengthening its competitive position in the rapidly growing voice AI market. The deal reflects the intensifying competition among voice AI platforms.
|
|
Rivian expands joint venture with VW to include AI assistant
Rivian is expanding its $5.8 billion joint venture with Volkswagen to now include the development of an AI assistant for vehicles. The company has not disclosed whether the financial value of the joint venture will increase as a result of the expanded scope. The move signals a growing emphasis on AI-driven in-car experiences within the EV industry.
|
|
The NYT Cooking app is testing AI recipe modifications
The New York Times is piloting an AI feature within its Cooking app that allows users to modify recipes by suggesting ingredient substitutions, equipment alternatives, and scaling adjustments. The feature was announced by NYT Cooking senior editor Genevieve Ko and has already sparked debate among the app's user base. The experiment reflects a broader trend of media companies embedding AI tools into their consumer products.
|
|
Granola crunches up a dictation feature to compete with Wispr
Meeting notetaker app Granola is adding a dictation feature to its product, directly competing with Wispr—which recently added meeting notes to its own dictation app. The tit-for-tat expansions between the two startups illustrate the intensifying competition in the voice AI productivity space. The moves suggest the boundaries between dictation and meeting notes apps are rapidly blurring.
|
|
Amazon updates Alexa to help you manage shopping lists and subscriptions and update delivery instructions
Amazon has rolled out new updates to Alexa that enable users to create and manage shopping lists, handle subscriptions, and update delivery instructions using voice commands. The update adds more practical utility to Alexa's day-to-day functionality as Amazon continues to evolve its voice assistant. The enhancements reflect Amazon's ongoing effort to keep Alexa relevant in an increasingly competitive AI assistant landscape.
|