Apple is losing its edge?

Plus, šŸŽ¤ Build Promo Pages with Just Your Voice, Using Gemini 2.5 + Replit, UK Fast-Tracks Housing Plans with Google’s Gemini-Powered AI Tool, and more...

Hola DecoderšŸ˜Ž

If someone forwarded this to you and you want to Decode the power of AI and be limitless, then subscribe now and Join Decode alongside 30k+ code-breakers untangling AI.

šŸ‘€ Apple’s AI Whisper: Quiet Power Moves at WWDC 2025

Apple’s WWDC 2025 keynote leaned into OS polish, design refinements, and privacy, but stayed unusually low-key on the AI front. In a year where competitors are screaming AI from the rooftops, Apple’s strategy felt more like a silent handshake than a grand reveal.

  • Subtle AI Feature Drops: Live Translation now works across Messages, FaceTime, and calls, processed fully on-device for privacy. A new Visual Intelligence layer lets users analyze on-screen content, search for similar items, and even ask ChatGPT-style questions about images.
  • Smarter Automation, Developer Access: The Shortcuts app gains AI-powered suggestions and can now pull from ChatGPT for workflow automation. Apple also opened up access to its on-device model for developers, enabling AI integration without cloud costs.
  • AI Coaching Enters the Gym: Workout Buddy on Apple Watch uses real-time biometric data and user history to deliver personalized, voice-guided coaching, bringing AI quietly into daily routines.

Conclusion: Playing the Long AI Game

Apple’s updates show it’s building AI from the inside out, but without the flash. Compared to rivals chasing headlines, this felt like a strategic infrastructure year. The question is: Will quiet confidence pay off in an AI-saturated world?

Together with Modash

šŸš€ Scale Partnerships Without Clunky Tools, Missing Data, or Wasted Spend

Most influencer marketing tools simply aren’t built for scale, let alone expansion into new markets. Even worse? Limited reach, irrelevant creators, inefficient workflows, and missing performance data.The result: costly misfires and underwhelming campaigns.

Modash changes everything.

āœ… Discover niche-fit creators from a database of 250M+ profiles
āœ… Reach out, ship products, and track performance, without ever switching tabs āœ… 14-day free trial and flexible monthly plans designed for modern growth teams

If you’re serious about expansion, the right system isn’t a ā€œnice to haveā€; it’s non-negotiable.

Just ask brands like Stanley, Loop Earplugs, and Regal Rose, who’ve scaled smarter with Modash.

šŸŽÆWant the exact blueprint for entering new markets or going hyper-local with creators?

Join Modash’s free webinar on June 18th with Piia ƕunpuu, Global Influencer Marketing Manager at Bolt. She’ll share how her team stacks creators, targets specific regions, and scales smarter without wasting budget.

Save your spot now and access the system behind top-performing brands!

Can’t make it live? No worries, register anyway to receive the full recording.

šŸŽ¤ Build Promo Pages with Just Your Voice, Using Gemini 2.5 + Replit

Say the offer. Get a page. No dev team required.

Launching a flash sale or product drop? You don’t need a landing page designer anymore. With Google’s open-source Gemini 2.5 agent and Replit, you can now generate full HTML/CSS promo pages using simple voice prompts like: ā€œCreate a weekend sale banner for 30% off serums.ā€

Here’s how to build your AI-powered Promo Page Builder in minutes:

šŸ›  Step-by-Step Guide

1. Import the Gemini 2.5 Repo in Replit

→ Go to Replit

→ Click ā€œImport from GitHubā€

→ Paste this repo: https://github.com/google-gemini/gemini-fullstack-langgraph-quickstart

2. Run Setup Commands

→ Terminal: pip install -e ./backend && npm install --prefix frontend make dev

3. Set Up Your API Key

→ Rename .env.example to .env

→ Paste your Gemini API key

4. Replace Frontend Config

→ In frontend/vite.config.ts, paste this config

5. Add the Promo Page Generator Node

→ Modify LangGraph to include a node that converts marketing prompts into HTML/CSS using Gemini’s web dev capabilities.

6. Use Voice to Prompt

→ Add a speech-to-text layer (e.g., Web Speech API)

→ Speak your prompt: ā€œPromo page for summer deal: 20% off all bundlesā€

The result? A clean, custom HTML/CSS page generated in seconds, ready for ads, email, or Shopify. Build once. Reuse for every campaign.

šŸ—ļø UK Fast-Tracks Housing Plans with Google’s Gemini-Powered AI Tool

The UK government is partnering with Google to modernize its notoriously slow planning system using Gemini AI. Their new tool, ā€œExtract,ā€ promises to slash bureaucracy and unlock faster housing and infrastructure approvals nationwide.

1. Multimodal file conversion at scale: Extract uses Gemini to read, interpret, and digitize legacy planning files, including blurry scans, handwritten notes, and maps. This turns inaccessible archives into searchable, structured data.

2. From 2 hours to 40 seconds: The AI reportedly reduces what would take a planner 2 hours into just 40 seconds. It allows civil servants to skip tedious manual checks and focus on higher-value decisions.

3. Trials underway, full rollout in 2026: Several UK councils are currently piloting the tool, with a nationwide rollout expected by Spring 2026. It’s a key step toward hitting the government’s 1.5 million home target.

4. Human-led planning, AI-backed efficiency: The intent is not to replace planners, but to amplify their capacity and ease backlogs. It’s about pairing human judgment with AI-powered throughput.

Why it matters:

Governments often lag in adopting modern tools, but Extract shows how AI can quietly transform critical systems without flashy disruption. The biggest early AI gains might not be futuristic, but practical: digitizing paperwork, eliminating bottlenecks, and moving real-world infrastructure forward.

šŸ† Tools You Cannot Miss:

šŸ—ļø Rierino – Build and deploy composable digital solutions at scale. An enterprise-grade platform for future-ready teams.

šŸŽ™ļø Merlio – Talk, generate, and design with every major AI model in one place. A unified app for multimodal creativity.

🧠 Soul Machines Studio – Create interactive, autonomous AI assistants. Bring lifelike avatars to real-world roles.

šŸŽ¶ freebeat AI – Turn music and ideas into viral videos with one click. Your next content hit is AI-generated.

šŸ“š Doc2Exam – Turn any study material into live, interactive exams. From passive reading to active learning in minutes.

šŸš€ Quick Hits

🧪 Why wait for campaign results when you can predict them? Neurons gives you instant insight into how your ad will perform, from attention heatmaps to memory scores. It’s like pre-launch A/B testing, with neuroscience. Book your free demo and pre-test every creative.

šŸ‹Chinese AI chatbots like Qwen, Doubao, and Yuanbao have paused image-based question photo features during June exams to stop cheating. Over 13 million students are taking the tough ā€œgaokaoā€ college test

šŸŸQualcomm will acquire U.K. chipmaker Alphawave Semi for $2.4B to boost its data center business. The deal, set to close in Q1 2026, strengthens Qualcomm’s AI and cloud connectivity push.

šŸ—£ļøOpenAI upgraded ChatGPT’s voice mode for paid users, making speech more natural with improved emotion and cadence. It now supports continuous translation, though minor audio glitches and hallucinations remain.

šŸ’µOpenAI says it has reached $10 billion in annual recurring revenue, doubling from last year. The growth comes from ChatGPT, API, and business tools, with 500M weekly users and 3M+ paying businesses.

Thanks for Decoding with us🄳

Your feedback is the key to our code! Help us elevate your Decode experience by hitting reply and sharing your input on our content and style.

Keep deciphering the AI enigma, and we'll be back with more coded mysteries unraveled just for you!