• The Shift
  • Posts
  • Gemini makes your Camera Smarter

Gemini makes your Camera Smarter

Plus, ⚡ Launch a Live Landing Page in Minutes Using Gemini Canvas, Shopify’s “AI-First” Hiring Rule, and more!


Hello there! Ready to dive into another upgrading, Mind-boggling, and Value-filled Shift?

Today we have:

📸 Google’s AI Gets a Visual Upgrade

⚡ Launch a Live Landing Page in Minutes Using Gemini Canvas

🤖 Shopify’s “AI-First” Hiring Rule

🏆 Tools and Shifts you Cannot Miss

📸 Google’s AI Gets a Visual Upgrade

Google is doubling down on visual intelligence with major updates to its AI Mode and Gemini Live. These new features allow users to ask smart questions about images, both from their camera and screen, merging visual input with conversational AI for deeper, more intuitive results.

The Decode:

1. AI Mode gets multimodal image understanding - AI Mode now lets users upload or snap photos and ask complex questions about the contents. Powered by Google Lens, it interprets object relationships, shapes, and context to deliver nuanced, contextual answers. You can even ask follow-up questions to refine the results further.

2. Gemini Live now works on camera and screen - On Pixel 9 and Galaxy S25 phones, Gemini Live allows users to point the camera at real-world objects or share screens to get AI-powered answers. Whether it’s identifying fish at an aquarium or comparing jackets online, Gemini Live processes it all in real time.

3. Rolling out widely—but with limits - AI Mode is now available to millions of Google Labs users beyond AI Premium subscribers. Gemini Live is rolling out to select Android devices but requires a Gemini Advanced subscription. It supports 45 languages, though education and enterprise accounts are excluded.

Google is turning smartphones into intelligent assistants that can see, interpret, and respond. For marketers, educators, and users alike, this opens new dimensions for search, shopping, learning, and content discovery.


⚡ Launch a Live Landing Page in Minutes Using Gemini Canvas

No code? No problem. With Gemini 2.5 Pro + Canvas, you can now build and deploy a fully functional landing page — powered entirely by AI — in under 10 minutes, completely free. Perfect for launching MVPs, side projects, or testing new ideas on the fly.

🔧 Step-by-Step Guide

1. Fire Up Gemini

Head to Gemini, select 2.5 Pro (experimental), and activate the Canvas toggle just beneath the prompt bar.

2. Drop This Prompt

“Act as a product designer for a fitness tech brand. Create a responsive landing page for an AI-powered workout planner. The page should explain how users can get custom workout plans in seconds, feature visual mockups, client reviews, pricing cards, and a compelling CTA. Use HTML, CSS, and minimal JS. Keep it modern, clean, and conversion-optimized.”

3. Let Gemini Work Its Magic

In seconds, Gemini will generate your landing page and show you a live preview. You can tweak design, copy, or layout with follow-up prompts.

4. Save Your File

Click to view the code, copy it, and paste into a new file named index.html inside a folder called “WorkoutPage” (or any name you like).

5. Deploy with Cloudflare

Create a free Cloudflare Pages account → select Workers & PagesCreate ProjectUpload FolderDeploy.

🌍 Go Live Instantly

You’ll receive a free public URL. Or connect your own domain for a polished finish. It’s never been easier to launch an idea with speed and style.

🤖 Shopify’s “AI-First” Hiring Rule

Shopify CEO Tobi Lütke just made AI a gatekeeper for hiring. In a bold internal memo, he declared that no team can request more headcount or resources without proving AI can’t do the job first, making AI usage a baseline expectation for all employees.

The Decode:

  1. AI Before Headcount - Before any team can ask for new hires or budget, they must show why AI can’t handle the task. Lütke encouraged teams to envision their work if autonomous AI agents were already teammates.

  2. Reflexive AI Usage is Mandatory - Using AI effectively is now a fundamental expectation at Shopify. Proficiency with AI tools is part of performance and peer reviews, positioning it as a baseline skill for every employee.

  3. Global Disruption Adds Pressure - A new report from the UN’s Trade and Development organization warns that AI could disrupt over 40% of jobs globally. Lütke’s policy reflects this tension—prioritizing efficiency while raising deeper concerns about employment and automation ethics.

  4. Fewer Humans, More Tech - Shopify already cut 20% of its workforce last year and trimmed support roles in January. Lütke joins leaders like Sebastian Siemiatkowski, Klarna’s CEO, who says AI replaced 700 roles—signaling a broader shift toward leaner, AI-driven teams.

Shopify is drawing a line in the sand: human hiring must now justify itself against machine alternatives. As AI becomes embedded into job expectations and operational strategy, this memo could signal how future workplaces operate—where proving you're better than AI becomes part of the job.


🏆 AI Tools for the Shift

🧠 MotionGet expert-level ad analysis in minutes using AI agents trained by real performance marketers. Get your early access now!

🏆 SocraTurn your goals and challenges into focused execution with AI support and build momentum and crush your to-dos, one step at a time.

💊 DrugCardMonitor drug safety and streamline pharmacovigilance with AI.
Simplify compliance and protect patient outcomes faster than ever.

📸 DreamShootAIUpload a selfie, get a studio-quality photoshoot with AI.
No need for expensive gear—look amazing from your couch.

💼 JobbyoLand more interviews with AI-enhanced job tools and smart suggestions.
Stand out in every application with personalized support.


🚀Quick Shifts

💭 Major publishers, including NYT and Vox Media, launched a “Support Responsible AI” campaign urging U.S. lawmakers to regulate AI content usage. The initiative calls for fair compensation, attribution, and safeguards to prevent unauthorized use of journalistic content in AI training models.

📚 Google's DeepMind is reportedly enforcing year-long noncompete agreements on select AI researchers in the U.K., effectively paying them to avoid joining competitors like OpenAI. While meant to retain talent, the move is leaving some frustrated over stalled progress.

🚩 Amazon has upgraded its AI video model, Nova Reel, to generate clips up to two minutes long with consistent multi-shot style. Available via AWS, the tool now includes a manual mode for more visual control—access requires approval.

🌟 Meta’s Llama 4 Maverick model impressed with top AI benchmark scores—but it turns out the tested version was a custom-tuned variant not publicly available. LM Arena has since revised its policies, raising concerns about fair benchmarking and developer trust.


That’s all for today’s edition see you tomorrow as we track down and get you all that matters in the daily AI Shift!

If you loved this edition let us know how much:

How good and useful was today's edition

Login or Subscribe to participate in polls.

Forward it to your pal to give them a daily dose of the shift so they can 👇

Reply

or to participate.