Good morning! Here's what's happening in AI today:
Google launches Gemini for Slides to build presentations automatically
Claude can now answer questions using the Anthropic Economic Index
HeyGen launches Companion Mode for directing videos step by step
How to teach Claude to not sound like AI
And 4 new AI tools worth trying today
AI PRODUCTIVITY
Google rolled out Gemini inside Google Slides, letting it read your own Docs, Sheets, and PDFs and generate a presentation directly from them instead of starting from a blank deck.
Gemini pulls the actual content and formatting cues from files you already have, so the first draft of the deck reflects real numbers and real language from your documents, not generic placeholder text.
Every slide it generates stays fully editable afterward, you can move elements, rewrite text, or swap layouts the same way you would in a deck you built by hand.
The feature works inside Google Slides right now, no separate tool or export step required to get from source documents to a working draft.
Most presentations start with pulling numbers from a spreadsheet, quotes from a doc, and then manually laying all of it out slide by slide. Gemini does that first pass for you, using files you already had open anyway. It doesn't replace editing the deck, it replaces staring at a blank one.
AI RESEARCH
Anthropic connected Claude directly to the Anthropic Economic Index, its public dataset tracking how AI use shows up across different jobs and industries.
You can ask which occupations show the heaviest AI usage and get an answer sourced from the actual Index data, not a general estimate pieced together from training knowledge.
Questions about which specific tasks are getting automated pull from that same real dataset, so the answer reflects measured usage patterns rather than a guess.
The Index itself remains publicly available separately, so Claude surfacing it in conversation doesn't limit what you can dig into on your own.
Reading through an economic research report to find one specific stat is slow, and most people never do it. Wiring Claude directly to the Index means you can ask a specific question and get a specific, sourced answer in the format you actually asked for.
AI COMMUNICATIONS
HeyGen introduced Companion Mode, which walks you through building a video in stages, pitching angles and sketching a storyboard, instead of generating one finished clip from a single prompt.
The tool proposes a few different creative directions for your video upfront, so you're choosing between real options before any footage gets generated.
It sketches out a storyboard for you to review at each stage, giving you a checkpoint before committing to the next part of the build.
Nothing renders until you've approved each stage, so you're not stuck regenerating an entire video because one section didn't match what you wanted.
A single prompt into a video tool is a gamble, you either get what you pictured or you don't, and there's rarely a middle ground to adjust. Companion Mode adds actual checkpoints along the way, so you catch a wrong direction at the storyboard stage instead of after the whole clip is rendered.
HOW TO AI
The sentences are too balanced. Every paragraph runs the same length. Nothing sounds like a specific person said it.
Most people try to fix this by prompting harder. That doesn't work, because the problem isn't the prompt. It's that Claude has never actually been told what your writing sounds like, so it defaults to the safest, most neutral version of English it was trained on.
The fix isn't a better prompt. It's a file, one you build once, that Claude reads before it writes anything for you again.
Step 1: Interview Yourself
Open a new Claude chat and paste this:
I want to create a writing style guide that captures exactly how
I write, so I can give it to AI and have it produce content that
sounds like me. Interview me with specific questions about:
1. My sentence length preferences and rhythm
2. Words and phrases I commonly use
3. Words and phrases I never use or hate
4. How I open pieces, do I use questions, bold statements, stories?
5. How I close pieces
6. What tone I aim for (direct, warm, funny, formal, conversational?)
7. What I think makes writing good vs. bad
8. Examples of writing I admire
Ask me one section at a time. After I answer everything,
create a clean markdown file I can download and use as a
permanent reference.Answer specifically. "I don't like formal language" gives Claude no actual rule to check against.
Step 2: Choosing Your Banned Words
Start with the words that are already common AI tells, then add every word or phrase that makes you cringe in something AI wrote.
Step 3: Write Your Positive Rules
Banning words only covers what to avoid. Cover sentence style, structure, openings, and tone, customized to describe you specifically. If this file could have been written by anyone, it isn't finished.
Step 4: Save It and Make It Automatic
Save the file as writing-style.md. Upload it to every Claude project where you produce content, and add one line: read the guide before writing anything, and treat it as not optional.
Step 5: Set Up a Folder System, Not Just a File
Keep three folders wherever you work with Claude: ABOUT ME, TEMPLATES, and OUTPUTS.
Step 6: Write With One Prompt, Every Time
Once the structure is in place: "I want to [TASK] for [SUCCESS CRITERIA]." After the first draft comes back, ask Claude to audit it against writing-style.md. Repeat until the flagged patterns are gone. The full guide is here.
P.S. You can access all my prompts, workflows and AI trainings if you upgrade here
Anthropic launched the Claude Security plugin for Claude Code in beta, letting you scan for vulnerabilities before you commit or run a full codebase scan from the terminal.
Microsoft Research fully open sourced MagenticLite's models, with MagenticBrain and Fara 1.5 now live on Hugging Face with open weights.
Lovable updated its apps to connect with each user's own tools, giving every person a personalized build based on their own data, workflows, and permissions.
🎯 Manus Plan Mode: Turns a rough idea into a plan you approve before it builds.
🎨 Qwen-Image-3.0: Generates near photographic images with legible text and layouts.
🔍 Firecrawl /search: Returns exact excerpts that answer your query, free by default.
🎤 ElevenLabs Vocals: Generate original songs in your own voice or a library one.
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
✓ Full archive of premium guides with ready-to-use prompts
✓ Structured AI courses (step-by-step, start-to-finish)
✓ Every upcoming premium tutorial