Claude Sonnet 4.6: 1M Context Window, Stronger Coding, and Near-Opus Performance - Anthropic has released Claude Sonnet 4.6, introducing a 1-million-token context window and major upgrades in coding, computer use, and long-context reasoning while keeping pricing unchanged. The model approaches Opus-level performance across several benchmarks and is now the default across Claude plans.

Why it matters: Advanced AI capabilities are rapidly moving into lower-cost models, making large-scale enterprise deployment and agentic workflows increasingly practical.

Perplexity Rejects AI Advertising to Compete on Trust and Subscription Search - Perplexity is stepping away from advertising, arguing that ads could make users “doubt everything” in AI-generated answers. The decision puts the company at the center of a growing industry divide as OpenAI experiments with ads, Anthropic rejects them, and Google blends advertising into AI search results.

Why it matters: The future of AI search may depend less on which model is smartest — and more on which business model users trust.

OpenAI Has Poached Instagram’s Celebrity Whisperer - OpenAI has reportedly hired longtime Instagram executive Charles Porch to help guide consumer growth and product engagement, signaling a deeper push into mainstream social-style adoption beyond enterprise AI tools. The move suggests OpenAI is increasingly focused on how everyday users discover, interact with, and stay engaged with AI products — not just how powerful the models are.

Why it matters: The next phase of AI competition may be driven less by model capability and more by product experience, distribution, and user retention — areas where social media expertise could become a strategic advantage.

World Labs lands $1B, with $200M from Autodesk, to bring world models into 3D workflows - AI startup World Labs, founded by Fei-Fei Li, has secured $200 million in funding backed by Autodesk to integrate “world models” — AI systems that understand physical environments — directly into 3D design and creative workflows. The technology aims to enable AI systems that can reason about space, motion, and real-world environments rather than generating static outputs.

Why it matters: World models represent a major step toward AI that can simulate reality itself, potentially transforming industries like architecture, robotics, gaming, and digital production by allowing AI to operate inside spatial environments instead of text interfaces.

ChatGPT Atlas hints at AI becoming the next browser layer - A post from OpenAI product leader Adam Fry previewed ChatGPT Atlas, a macOS browsing interface that integrates ChatGPT directly into web navigation through a persistent side panel. The feature allows users to ask questions about any webpage, summarize content, and organize research without leaving their browsing session. This week’s Atlas update introduces revamped tab search and a new “auto organize” tabs feature, allowing users to remove duplicates, merge windows, or automatically group tabs using ChatGPT’s organizational logic. Users can access the update directly through the Atlas interface by selecting “update” in the top-right menu.

Why it matters: If AI becomes an embedded navigation layer within browsing rather than a separate destination, it could reshape search behavior, advertising models, and how users interact with information online — shifting AI from a tool users visit into an assistant that actively helps manage the web itself.

Grab a product photo first — Runway will use it for the start frame. Don’t have a product? Use a software mockup or ask Gemini to generate one for the brand

Drop the photo into any AI, prompting: "Write me a brief video prompt to use in Runway to generate a social media video based on this product photo. Add animations and/or camera movement if applicable"

Next, go to Runway > Tool > Video and add your starting frame and the AI-generated prompt

Rinse and repeat for the rest of your product photos. Try different combinations of text overlay and music when you post them to socials

Descript: Descript is an all-in-one audio and video editing tool that lets you edit media as easily as a text document, with AI features for transcription, overdubbing, and streamlined content production. We’re excited to invite you to join Descript’s Creator Plan and enjoy 50% off your first 2 months. Be sure to use our link to activate the discount! (As an affiliate partner with Descript, AiNews may earn a small commission if you sign up using our link. There’s no extra cost to you — it simply helps support our independent reporting and the work we do.)

Lindy: Text your AI assistant. Get Answers. Get things done. Lindy saves you two hours a day by proactively managing your inbox, meetings, and calendar, so you can focus on what actually matters.

AdScriptly: AI Ad Optimization for Predictable Revenue. Smarter Google & Meta ads, powered by offline business data.

GoAI: See the Opportunities Others Don't. Stop chasing the news. GoAI alerts you to high-conviction alpha opportunities and portfolio risks 24/7—before the market reacts.