- AI Toast
- Posts
- Google released Nano Banana 2
Google released Nano Banana 2
Plus: How to schedule a task with Claude Cowork
Poonam Soni
February 27, 2026
In partnership with
Welcome to AI Toast!
Here’s what’s on the menu today:
Google releases Nano Banana 2 🤩
Anthropic Sets Limits on Military AI
Superhero Tools for Chatbots of 2026
Step by step tutorial on how to scheduled tasks in Claude Cowork
Quick Toasts: Fast food just got an AI supervisor.
Total read time: About 5 minutes, perfect for a quick coffee break.
Image: Google blog
Google didn't just update their image model, they rebuilt it. And this one is worth paying attention to. It's their most advanced image generation and editing model yet built on top of Gemini 3.1 Flash. It generates high-quality images faster, handles complex prompts better, and costs less to run at scale.
It taps into Gemini's world knowledge to create more realistic, detail-rich visuals even pulling from web images as reference.
Text inside images is now sharp and accurate and it works across multiple languages, so you can localize ads, posters, or any visual content without starting over.
You get full creative control: native aspect ratios including 4:1, 8:1 and 1:4, a new 512px resolution tier for faster outputs, and adjustable thinking levels so the model can reason through tricky prompts before rendering.
It keeps your subject looking consistent across different scenes useful for product shots, branded content, or anything that needs visual continuity.
You can try it as the new default in the Gemini app, Search (across 141 countries), and Flow + available in preview via GoogleAIStudio and Vertex AI.
Personal Take: This feels like the moment image generation gets serious for production teams. The combination of fast speed, real-world knowledge, and reliable text rendering removes a lot of the friction developers usually hit. If you're building anything visual at scale ads, UI, creative tools, this is worth testing right now.
Vibe code with your voice
Vibe code by voice. Wispr Flow lets you dictate prompts, PRDs, bug reproductions, and code review notes directly in Cursor, Warp, or your editor of choice. Speak instructions and Flow will auto-tag file names, preserve variable names and inline identifiers, and format lists and steps for immediate pasting into GitHub, Jira, or Docs. That means less retyping, fewer copy and paste errors, and faster triage. Use voice to dictate prompts and directions inside Cursor or Warp and get developer-ready text with file name recognition and variable recognition built in. For deeper context and examples, see our Vibe Coding article on wisprflow.ai. Try Wispr Flow for engineers.
Anthropic CEO Dario Amodei published a public statement this week explaining why the company is refusing certain requests from the Department of War, even if it means losing one of their biggest government contracts. It's a rare move for an AI company, and it's already sparking a lot of conversation.
Anthropic has been deeply embedded in US government systems, they were the first AI company to deploy models in classified networks, the National Laboratories, and national security agencies.
Claude is currently used for intelligence analysis, cyber operations, military planning, and more. This isn't a company that's been sitting on the sidelines.
But the Department of War is now demanding "any lawful use" access meaning no safeguards at all and Anthropic said no.
The two things Anthropic refuses to allow: mass domestic surveillance of American citizens, and fully autonomous weapons that remove humans from targeting decisions entirely.
The DoW has threatened to label Anthropic a "supply chain risk" a designation that has never been applied to an American company before and to use the Defense Production Act to force the removal of those safeguards.
Amodei pointed out the contradiction directly: they can't simultaneously be a national security risk and essential to national security.
Anthropic says if they're removed, they'll help with a smooth transition so no military operations are disrupted.
Personal Take: This is one of the most important moments in AI so far. Most companies quietly comply or avoid the conversation entirely. Anthropic is putting it in writing and making it public. Whether you agree with their position or not, it forces a real conversation about where the limits of AI in warfare should be and who gets to decide that.
Stop Drowning In AI Information Overload
Your inbox is flooded with newsletters. Your feed is chaos. Somewhere in that noise are the insights that could transform your work—but who has time to find them?
The Deep View solves this. We read everything, analyze what matters, and deliver only the intelligence you need. No duplicate stories, no filler content, no wasted time. Just the essential AI developments that impact your industry, explained clearly and concisely.
Replace hours of scattered reading with five focused minutes. While others scramble to keep up, you'll stay ahead of developments that matter. 600,000+ professionals at top companies have already made this switch.
Superhero Tools (Chatbots of 2026)
Gemini: Google's multimodal AI assistant that integrates deeply with Google Workspace, Search, and Maps to help you research, write, and get things done faster.
Claude: Anthropic's AI assistant known for nuanced reasoning, long document analysis, and safe, helpful conversations across writing, coding, and complex tasks.
ChatGPT: OpenAI's flagship conversational AI that handles everything from drafting emails and writing code to image generation and voice conversations.
Microsoft Copilot: Microsoft's AI companion built into Windows, Edge, and Office 365 that helps you summarize documents, generate content, and automate work across the Microsoft ecosystem.
Grok: xAI's chatbot integrated into X (Twitter) with real-time access to posts and trending topics, designed for witty, unfiltered, and up-to-date responses.
Perplexity: An AI-powered search engine that gives you direct, cited answers to questions by pulling live information from the web ideal for research and fact-checking.
How to schedule a task with Claude Cowork
Go to https://claude.ai/ in your web browser. Click "Sign Up" or log into your existing account and enter your credentials to access your dashboard.
Once logged in, navigate to the Cowork section from the sidebar or app launcher. Cowork is Claude's desktop tool designed for automating file and task management without needing to code.
Download and install the Claude Cowork desktop app if you haven't already. Follow the on-screen setup prompts, it connects directly to your Claude account automatically.
Open Cowork and go to the "Tasks" or "Automations" panel. Click "New Task" or "Schedule Task" to begin setting up your workflow.
Name your task and describe what you want Claude to do be specific. Example: "Every Monday at 9am, summarize my downloads folder and move PDFs to /Documents/Reports."
Set your trigger: choose a schedule (daily, weekly, custom cron) or a condition-based trigger (file added, folder change, time of day).
Define the action: describe the steps Claude should take in plain language moving files, renaming, summarizing, sending a report, etc. You can chain multiple steps together in one task.
Set any output preferences: where results should be saved, whether to get a notification or summary when the task runs, and what format outputs should take.
Hit "Save & Activate" to turn the task on. Claude will run it automatically based on your schedule. You can pause, edit, or delete tasks at any time from the task list.
Review task run history in the "Logs" section to check what was done, catch any errors, and refine your instructions if the output wasn't quite right.
Iterate your task descriptions for better results, the more specific your instructions, the more accurate Claude's actions. Duplicate tasks to create variations for different folders, projects, or time windows.
That's it! Most scheduled tasks are set up in under 3 minutes once you know what you want automated. If a task doesn't run as expected, your description was likely too vague, add more context about file types, folder paths, or exact actions and reactivate.
Quick AI News Bites
Samsung’s Galaxy S26 lineup launches with expanded AI features and multiple assistants Bixby, Gemini, and Perplexity all integrated into the new phones for smarter, personalized mobile AI.
OpenAI names London its biggest research hub outside the U.S., aiming to grow AI research and collaboration in Europe. The move feeds into Britain's push to cast itself as an "AI superpower" and a home for cutting-edge research at a time when governments are vying for investment from major model developers.
Burger King is testing an AI chatbot named “Patty” that lives in employees’ headsets to help staff and track service friendliness, powered by an OpenAI base model. Patty can respond to voice prompts, provide operational help (like recipes, inventory alerts, and out-of-stock updates), and listen for phrases like “please,” “thank you,” and “welcome” to give managers insight into overall service patterns though the company says it’s a coaching tool, not a performance tracker.
Boost revenue and gain new customers by partnering with us
Reach over 35K AI enthusiasts with your product.
Join our newsletter to connect with tech professionals, investors, engineers, managers, and business owners worldwide. DM now!
Got feedback, a story worth toasting, or a wild tech question? I’d love to hear from you; just reply, or find me on X.
Cheers,
— Poonam Soni