|
In this newsletter:
- Claude Voice Mode now runs on Opus and Sonnet models with app control
- OpenAI launches a small business program with ChatGPT work access & partner tools
- Runway Agent 2.0 turns one idea into a finished video in a single workflow
Plus, you’ll find new AI tools and this week’s top AI news headlines!
🎙️Claude's Voice Mode Now Runs on its Most Powerful Models and Controls Your Apps
Anthropic upgraded Claude's voice mode to run on Opus and Sonnet models, added the ability to take action in connected tools like Gmail and Canva, and expanded language support to 11 languages.
What's new:
- Available now in beta to all chat users on mobile, desktop, and web; works best on phone
- Claude Opus and Sonnet now available in voice mode alongside Haiku
- Switch models mid-conversation from the model picker
- Defaults to last model used in text chat so you can move between voice and text without starting over
- Voice mode can now take action in connected tools after asking for permission
- 11 languages now available: English, French, German, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese (Brazilian), Spanish Latin America, Spanish Spain
- Switch languages mid-conversation by telling Claude out loud or selecting from the language picker
- Free plan gets Haiku, one connected tool, all languages
- Paid plans get expanded models and all connected tools
What you can do:
- Practice for an important conversation and ask Claude for feedback on your communication style
- Talk through a client pitch and ask Claude to find gaps in your logic
- Brainstorm product roadmap ideas and ask Claude to validate them with competitive research
- Ask Claude to push a meeting on Google Calendar by 30 minutes
- Turn a conversation about a client pitch into a one-pager in Canva
- Ask Claude to summarize emails received today and draft responses to the most critical ones
Why it matters:
Voice mode going from Haiku to Opus and Sonnet is the difference between a quick lookup tool and an actual thinking partner you can have a real back-and-forth with out loud.
The more interesting unlock is tool access - being able to talk through a problem and then tell Claude to act on it without switching to text or touching your phone changes how useful voice AI actually is for getting work done, rather than just being a hands-free way to ask questions.
👀 Read more about Claude's voice mode update!
💼 OpenAI Launches Small Business Program with Training, Partner Tools, and ChatGPT Work Access
OpenAI launched a dedicated small business program giving owners access to ChatGPT Work powered by GPT-5.6, alongside free training, in-person events, and partner integrations with Dropbox, Shopify, Intuit, Slack, and more.
What's included:
- Virtual training webinars covering ChatGPT Work use cases across accounting, marketing, and ecommerce with specific prompts and demos
- In-person Small Business AI Academies across the US with guided instruction and hands-on exercises
- At last year's events, 78% of participants built a functional AI workflow in a single day, and 42% saved more than five hours a week
- New guides including customer stories, interactive guides you can upload directly into ChatGPT Work, and short-form videos
- Partner plugins, skills, and exclusive promotions from Dropbox, Shopify, Intuit, Slack, Atlassian, Wix
What ChatGPT Work can do for small businesses:
- Send a voice note and ChatGPT Work translates it into a Slack message sent to multiple channels at once
- Start every week with a site that updates in real time with market mentions, competitor performance, or product trends
- Evaluate current inventory and suggest new products or marketing campaigns to boost sales
- Create a project with customer reviews and build a training presentation celebrating wins and identifying improvements
- Powered by GPT-5.6, available on all subscription plans for businesses of any size
Why it matters:
OpenAI is going after the same market Anthropic targeted with Claude for Small Business in May, recognizing that small business owners spending hours on tasks like bookkeeping, marketing copy, and customer communications represent a massive untapped opportunity.
The in-person academy model is the more interesting bet here - getting local business owners in a room together to build real workflows in a day creates the kind of word-of-mouth adoption that no amount of online advertising can replicate.
👀 Read more about OpenAI’s small business program!
🎬 Runway Agent 2.0 Turns One Idea into Finished Video in A Single Workflow
Runway launched Agent 2.0, a creative AI agent that takes users from a rough idea through storyboarding, generation, and editing to a finished video, all within a single interface.
What's new:
- Available now for all Runway users
- Ranked first among six AI video agents on Physion Labs' Arc 1.0 benchmark for cinematic quality, coherence, and safety
- Bring video clips, audio, and images together into one timeline directly within the Agent interface
- Upload ad performance data as PDF; Agent analyzes it, proposes a new creative strategy, then creates the actual content from that brief
- Generates 25 images in different styles presented visually so you can react, discard, drag back into chat, and iterate
- Knows which models render motion well, which hold up for photorealism, which are faster but less precise, and selects accordingly
- Move between agentic generation and manual editing within a single session
Who it's built for:
- Marketers building social content, ad variants, or campaign assets
- People starting with a feeling or mood who need help brainstorming toward a finished result
- Cinematic creators turning existing content into video storytelling
- Anyone who knows their goal but needs something visual to react to before refining it
What's coming:
- Brand identity memory so Agent knows your style without briefing from scratch each time
- Batch creation for repeatable everyday tasks
- Task scheduling directly within Agent
- Deeper integration with other tools
Why it matters:
Runway is solving the problem that stops most people from finishing AI video projects - the constant context switching between generation, editing, and briefing tools that breaks creative momentum. Keeping everything inside one interface where the video stays front and center rather than buried in a chat thread is what separates a production workflow from a glorified chatbot.
The memory and brand identity features signal where this is heading: a creative tool that gets sharper the more you use it, rather than starting from zero every session.
👀 Read more about the Runway Agent 2.0 launch!
My Latest LinkedIn & X/Twitter Posts:
- ChatGPT vs Perplexity vs Grok vs Gemini, when to use each (view post)
- 10-step roadmap to build AI agents from idea to evaluation (view post)
- 3 expert ways you can use AI in your workflows (view post)
- 12 prompts that turn ChatGPT into your thinking coach (view post)
- 10 powerful ways to use ChatGPT’s Deep Research (view post)
In partnership with Kit:
As a creator, your time should be spent doing what you love, not juggling a dozen tools just to run your business.
Kit gives you everything you need in one place:
✅ Build and grow your email list (I use Kit for my newsletter)
✅ Easily monetize with paid newsletters and digital products
✅ Automate your emails with triggers and custom workflows
✅ Track what’s working and optimize with powerful insights
It’s not just email. It’s your entire creator business, simplified.
Join a thriving community of successful creators.
Use Kit for Free — and Start Building Smarter
New AI Tools to Boost Your Productivity:
- Neurohelper: Accesses multiple AI models in one workspace.
- Morable.ai: Turns goals into personalized action plans.
- Velo.ai: Creates product demos, training, and support videos.
- GhostChat: Combines live chat and AI customer support.
- Seofable: Writes and publishes SEO blog posts automatically.
- AdMakeAI: Creates advertising creatives.
- Crop.photo: Automates Shopify product image editing.
- AI Image Detector: Detects AI-generated images.
- image-generator: Generates images from text and reference images.
- Sited: Tracks your brand's visibility in AI search.
- Playyy: Creates and edits content in one AI canvas.
- Pounce: Finds potential customers on X and Reddit.
- SEOGrown: Finds why your brand isn't ranking in AI search.
- LumiMusic: Creates and edits songs from prompts or lyrics.
This Week's Top AI News Headlines:
- Google Introduces Selfie Video Sign-In, a New Identity Verification Method that Uses AI to Confirm Users are Real People Before Granting Access to High-Risk Google Accounts (View Article)
- OpenAI Expands ChatGPT Health to All U.S. Users, Giving Everyone Access to AI-Powered Health Information, Symptom Guidance, Medical Record Integration, and Personalized Wellness Support (View Article)
- Amazon Upgrades Alexa+ with New AI Capabilities that Let Users Control Multiple Smart Home Devices through Complex Natural Language Commands Instead of Individual Voice Requests (View Article)
- Black Forest Labs Unveils FLUX.3, a New AI Model that Can Generate both Images and 20-Second Videos with Synchronized Audio, with Early Access Initially Limited to Select Users (View Article)
- OpenAI Launches OpenAI Presence, a Real-Time AI Platform that Lets ChatGPT and Other AI Agents Understand User Location, Nearby Places, and Live Context to Deliver more Personalized Assistance (View Article)
- xAI Launches Workflows in Grok Build to Break Complex Tasks into Parallel Jobs Across Up to 1,024 AI Agents, Turning Natural Language Requests into Reusable Automated Workflows (View Article)
- Cursor, the AI-Powered Code Editor Built on Visual Studio Code, Introduces Router, a Smart AI Model Selection System that Automatically Chooses the Best AI Model for Each Coding Task to Improve Speed, Quality, and Cost Efficiency (View Article)
- OpenAI Brings ChatGPT Voice to the Desktop App, Letting Users have Natural Voice Conversations with ChatGPT Directly on Windows and macOS without Using a Web Browser (View Article)
- AI Researchers Dispute Claims that Moonshot AI, the Chinese Startup Behind the Kimi AI Assistant, Used Anthropic’s Fable to Train Kimi K3, its Latest Open-Weight AI Reasoning Model (View Article)
- ElevenLabs Adds References to Music v2, Letting Creators Guide AI-Generated Songs with Reference Audio to better Control Style, Mood, Instrumentation, and Overall Sound (View Article)
- Microsoft Introduces Hill Climbing, a Training Method that Helps its MAI Models Improve Through Iterative Self-Refinement, Powering Smarter Coding in GitHub Copilot and More Accurate Spreadsheet Assistance in Excel (View Article)
- OpenAI Expands ChatGPT into a Personal Health Assistant with New Medical Information Features, Health Record Integration, Symptom Tracking, and Wellness Support Tools (View Article)
- Meta’s New AI Ad Tool Automatically Paired an Optimistic Marketing Campaign with a Song About Human Extinction, Highlighting the Ongoing Risks of AI-Generated Creative Decisions (View Article)
- Google Says Gemini is Closing in on 1 Billion Users as it Expands Across Search, Android, Workspace, and Other Google Products to Reach more People than Any Standalone AI Assistant (View Article)
- Substack Launches an AI Detection Tool that Shows Readers How Much of a Newsletter was Written with Artificial Intelligence and How Much was Written by a Human (View Article)
- Anthropic Launches an Economic Index Connector that Lets Organizations Compare their Workforce with Industry-Wide AI Usage Trends to Measure AI Adoption and Identify Opportunities for Greater Productivity (View Article)
- Google Defends its Multi-Billion-Dollar AI Spending by Pointing to Rapid Cloud Growth, Saying Rising Demand for AI Services is Driving Revenue and Justifying Continued Infrastructure Investment (View Article)
Work With Me:
If you enjoy this newsletter, please forward it to your friends and colleagues.
Follow me on LinkedIn and X/Twitter to see my latest posts.
Have a wonderful week!
Andrew Bolis
|