Your Daily Best of AI™ News
🚨Amazon has issued legal threats to Perplexity over its agentic browsing tool that allegedly scrapes Amazon's site while disguising automated activity as human behavior. This marks the opening shot in what will become a massive battle over who controls web data access as AI agents proliferate.
The Big Idea
The voice interface shortcut everyone's ignoring

Typing is the bottleneck nobody wants to admit.
Average typing speed is 40 words per minute. Average speaking speed? 150 words per minute. Yet we're still hunt-and-pecking our way through PRDs, documentation, and planning sessions.
Voice tools like Wisprflow, SuperWhisper, and Monologue are changing this, and the early adopters are lapping everyone else.
How it actually works:
You talk through your idea -- out loud, stream of consciousness, while walking the dog or commuting. The AI transcribes it, structures it, and hands you back a formatted document.
I watched someone map out an entire SaaS product in 23 minutes of voice notes. Turned into a PRD that would've taken 3+ hours to type.
The workflow is dead simple: talk, transcribe, refine.
Why this hasn't caught on yet...
... is mostly social. People feel weird talking to their computer. Offices aren't set up for it. And there's a learning curve to "thinking out loud" coherently.
But once you get past that awkwardness? It's absurdly faster.
The best use cases aren't replacing typing entirely -- it's capturing thoughts that would never get written down because typing friction is too high.
Planning documents. Feature brainstorms. Bug reports. User research notes. All the stuff that lives in someone's head because "I'll write it up later" (which means never).
Voice tools make "later" become "now."
The tools are maturing fast...
Wisprflow integrates directly into Notion and Google Docs. SuperWhisper runs locally (nothing leaves your device). Monologue specializes in technical jargon and preserves your speaking style instead of over-polishing.
Some teams are doing "voice-first planning" -- entire PRD sessions conducted verbally, AI listening and structuring in real-time.
The limitation is still editing. You'll always need to refine what comes out. But starting from 80% done via voice beats starting from a blank page typing.
What's next: Expect voice interfaces to become standard in AI coding tools. Cursor and Replit are both testing voice commands for code generation. "Build me a login form with email validation" while your hands stay on the keyboard for other work.
BTW: The voice-to-text accuracy on proper nouns and technical terms has gotten scary good. I dictated "Supabase row-level security policy" and it nailed it first try. Two years ago that would've been gibberish.

How To Build Apps Using AI (No Coding Required)
There is no better time to learn how to use AI to build solutions businesses want. I’m hosting a workshop that will show you, step-by-step, how to build tools and solutions using AI without having to know how to code.
If you’re interested, reply back to this email with the keyword BUILD and we’ll send you more info.
We are disclosing the dates soon. So if you’re interested make it known.
Today’s Top Story
Google's satellite gambit to solve AI's infrastructure crisis

The Recap: Google has unveiled Project Suncatcher, an ambitious moonshot to deploy AI chips on solar-powered satellites orbiting Earth. The project reveals just how desperate the infrastructure bottleneck has become for frontier AI development—when one of the world's largest tech companies seriously considers launching data centers into space, the on-ground limitations are clearly reaching critical mass.
Unpacked:
Space-based AI infrastructure would harness continuous solar energy without the constraints of terrestrial power grids, eliminating one of the largest bottlenecks in AI scaling.
The project underscores how hyperscalers are running out of conventional options—when billions in new data centers can't keep pace with compute demand, orbital alternatives start looking viable.
Google's willingness to pursue this technically complex and capital-intensive approach signals that AI infrastructure constraints may be more severe than publicly acknowledged by the industry.
Bottom line: Project Suncatcher represents a pivotal moment where AI infrastructure needs have outpaced Earth-bound solutions. The question isn't whether this specific moonshot succeeds, but what it tells us about the desperation driving frontier AI labs to explore increasingly extreme options to maintain their competitive edge.
Other News
Shopify reports AI traffic up 7x and AI-driven orders up 11x since January—early proof that AI agents are becoming a meaningful commerce channel, not just hype.
Sequoia names Alfred Lin and Pat Grady as co-stewards as Botha steps down after just 3 years—rare leadership transition at tech's most influential firm signals generational shift.
NVIDIA and Qualcomm join $850M+ push to build India's deep-tech ecosystem—chip giants making strategic bets on India as AI's next manufacturing and talent hub.
People Inc. licenses content to Microsoft's Copilot as Google traffic drops—legacy media pivoting from search SEO to direct AI licensing deals to survive.
Stability AI wins against Getty in UK court, but ruling avoids key copyright questions—landmark AI copyright case ends in legal limbo, leaving industry without clarity.
AI Around The Web
Coca-Cola’s annual Christmas advert is AI-generated again this year.
The company says they used even fewer people to make it — “We need to keep moving forward and pushing the envelope… The genie is out of the bottle, and you’re not going to put it back in”
— #DiscussingFilm (#@DiscussingFilm)
10:12 PM • Nov 3, 2025
Here's a quick 3 min tutorial on the absolute easiest way to use Wan Animate 2.2, Seedream and Elevenlabs voice2voice together to "remote control" any AI character in super high fidelity all inside of a single chat interface
— #fabian (#@fabianstelzer)
2:45 PM • Nov 5, 2025
Test Your AI Eye
Can You Spot The AI-Generated Image?


Prompt Of The Day
Copy and paste this prompt 👇
"Analyze [business], [product/service] and [product/service features]. Generate [number] pricing options for [product/service] along with the features that should give great value for the options. Name the pricing options with unique and simple words.
Business = [Insert Here]
Product = [Insert Here]
Product features = [Insert Here]
Number = [Insert Here]"P.S. Reply back to this email and let us know what area of AI you are struggling with the most. We are going to be hosting some free trainings and want your input.
Best of AI™ Team
Was this email forwarded to you? Sign up here.
