Good morning! Here's what's happening in AI today:
Grok releases speech-to-text inside Grok Build
Higgsfield launches an AI tool that turns any topic into a fully narrated documentary
Google brings Ask Gemini and AI Overviews to the Drive mobile app on Android and iOS
How to switch from ChatGPT to Claude
4 new AI tools worth trying today
AI TOOL
xAI activated speech-to-text inside Grok Build, letting developers dictate prompts directly to their coding agents using /voice or ctrl + space.
Use /voice or hit ctrl + space to start recording, and your spoken words are transcribed directly into the prompt field without switching apps or tools.
The feature is powered by Grok Voice and works inside the Grok Build environment where your coding agents already live.
You can speak naturally and the transcription appears in real time, so you can review before sending to your agent.
This removes the need to type out long, detailed coding instructions, which matters when you are iterating fast or working through complex multi-step tasks.
Typing detailed instructions to a coding agent slows you down. Voice input removes that friction entirely. For developers who already live in Grok Build, this means faster iteration without breaking flow. Speaking a prompt is faster than writing one, and that adds up across a full day of building.
AI VIDEO
Higgsfield introduced Higgsfield Explainer, a tool that researches any topic automatically, narrates it in any language, and renders a faceless documentary up to 10 minutes long in a single run.
Type any topic and the tool handles research, scripting, narration, and video rendering from start to finish with no manual editing required.
Videos can be narrated in multiple languages, so you can produce localized content for different audiences without recording anything yourself.
Each video renders up to 10 minutes in one run, which covers most short-form educational and explainer content without stitching clips together.
Higgsfield Explainer is available on the Higgsfield platform, via MCP, and on Supercomputer, powered by Claude Fable 5 and Gemini Omni Flash.
Faceless video is one of the fastest-growing content formats, but production has always been the bottleneck. Higgsfield Explainer takes you from a topic to a finished narrated documentary in one step. For educators, content creators, and anyone building an audience without showing their face, the production ceiling just got a lot higher.
AI PRODUCTIVITY
Google launched Ask Gemini and AI Overviews to the Google Drive mobile app for Android and iOS.
Ask Gemini lets you search and ask questions about any file in your Drive directly from your phone, without opening the document first.
AI Overviews surface an instant summary of a file the moment you tap it, so you can confirm it is the right one before reading the whole thing.
Both features work across Docs, Sheets, Slides, and PDFs stored in Drive, not just native Google file types.
The update brings capabilities that were previously desktop-only to the iOS and Android apps, so your full Drive becomes searchable and queryable from your phone.
Most people use Drive on mobile just to find and open files. With Ask Gemini built in, your phone becomes a place where you can actually work with your documents. You can query files, get instant summaries, and stop wasting time opening the wrong thing. This is one of the most practical AI additions Google has made to its mobile suite.
HOW TO AI
Most people who try switching fail for one reason: they bring their ChatGPT workflow into Claude and expect the same results. Claude was built around Projects, Skills, Connectors, and memory that loads automatically. Using it like ChatGPT misses the point entirely.
Here is the switch in four steps.
Step 1: Export your ChatGPT history
Go to Settings → Data Controls → Export Data in ChatGPT. You will receive an email with a zip file. Inside is chat.html, your entire conversation history. Also copy your Custom Instructions from Settings → Personalization before you close ChatGPT.
Step 2: Bring your history into Claude
Create a new Project in Claude. Upload chat.html to Project Knowledge. Every conversation you have ever had in ChatGPT is now permanent context that loads automatically into every chat inside that project.
Step 3: Transfer your memory
Go to Settings → Capabilities → Memory. Claude pulls your preferences from ChatGPT directly. Then take your Custom Instructions from Step 1, paste them into Settings → Profile → Instruction, and rewrite them as direct instructions rather than descriptions.
Step 4: Set up what ChatGPT does not have
Type /skill-creator in any Claude chat and build a skill for one recurring task. Then go to Settings → Connectors and connect the tools you use most. Add Project Instructions inside your project for standing context that never needs to be repeated.
The full guide is available for paid subscribers.
✓ Full archive of premium guides with ready-to-use prompts
✓ Structured AI courses (step-by-step, start-to-finish)
✓ Every upcoming premium tutorial
Anthropic opened registration for Built with Claude: Life Sciences, a global virtual hackathon running July 7-13 where builders can use Claude Science and Claude Code for a shot at $100K in credits.
Browser Use released CLI 3.0, a major update that lets any AI model control a real browser, now 6x smaller and using fewer tokens than before, with support for both cloud browsers and local Chrome.
Magnific launched an MCP integration that runs its full creative suite inside ChatGPT, letting you generate images and video, upscale, edit, and reframe without leaving the chat.
🔧 Grok Build: Code with AI agents using your voice, now with speech-to-text built directly into the Grok Build environment.
🎬 Higgsfield: Turn any topic into a fully narrated, research-backed AI documentary up to 10 minutes long, no editing required.
🖼️ Magnific: Generate images and video, upscale, and edit inside ChatGPT using Magnific's MCP integration.
🎞️ CapCut: Access Seedance 2.0 4K AI video generation across CapCut's app, web, and desktop platforms.
THAT’S IT FOR TODAY
Thanks for making it to the end! I put my heart into every email I send, I hope you are enjoying it. Let me know your thoughts so I can make the next one even better!
See you tomorrow :)
- Dr. Alvaro Cintas