$130 Billion in One Quarter

Welcome back to The Shift AI. Let’s get straight to what matters in AI today…

Today we have:

💰 Big tech just reported earnings, and AI is selling faster than they can build

🚀How to fix the technical issues hurting your AI search visibility with Semrush

🍳 A food AI startup claims it just had its Chatgpt moment

🔨Tools and Shifts you cannot miss

💰 Big Tech Just Reported Earnings and AI Is Selling Faster Than They Can Build

Microsoft, Google, Meta, and Amazon all reported Q1 earnings this week. The story was the same across all four, AI demand is outpacing their ability to build infrastructure fast enough to serve it.

The Shift:

Microsoft Is Running at $37B in AI - Microsoft's AI business hit a $37 billion annual run rate, up 123% in a year, and Copilot now has 20 million paid enterprise seats. Accenture alone signed up 740,000 in one deal.

Google Said the Quiet Part Out Loud - Google Cloud grew 63% to $20 billion, with Gemini AI products up nearly 800% year over year. Pichai admitted revenue would have been even higher if Google could build data centers fast enough.

AWS Just Had Its Best Quarter in Years - AWS grew 28%, its fastest in 15 quarters. Amazon's custom chip business crossed $20 billion, and OpenAI and Anthropic together committed gigawatts of compute capacity on Amazon's Trainium chips.

The Spending Numbers Are Hard to Comprehend - All four companies spent roughly $130 billion on infrastructure in a single quarter, nearly double a year ago. Google is now guiding to $190 billion in capex for 2026 and says it will go higher in 2027.

Meta grew 33% to $56.3 billion and raised its own capex guidance to $145 billion. Microsoft has $627 billion in signed but undelivered contracts. Google's cloud backlog doubled to $462 billion in one quarter. That is over a trillion dollars in commitments they cannot yet fulfill.

Together with Ghost

Your agent needs more than 2 projects

You prompt. The agent builds. Then it asks for a database.

Ghost is postgres made for this. Spin one up in seconds. Fork it like a branch. Delete it when you're done. Pay nothing when it's idle.

Your agent gets full sql, mcp support, as many databases. No dashboards. No provisioning. No forgotten dev databases draining your card at month end.

Build a weekend app. Fork the schema three different ways. Throw two of them out. Ghost doesn't care. The next prompt can spin up a fresh one.

You're already vibe-coding the app. Stop wiring up the backend.

Unlimited databases. Unlimited forms. 100 compute/hrs a month. 1tb of storage. Free.

Download Ghost

🚀How to Fix the Technical Issues Hurting Your AI Search Visibility With Semrush

Great content won't show up in ChatGPT if your site has technical blockers. Here's how to find and fix them.

Step 1: Run an AI visibility audit: Enter your domain into Semrush's AI Visibility tool. It surfaces the specific technical issues stopping you from being cited in AI responses.

Step 2: Fix your site structure: AI models pull from sources that are clearly structured and easy to parse. Semrush flags weak hierarchy, missing headers, and fragmented content that AI skips over.

Step 3: Strengthen authority signals: Semrush identifies where your domain authority, backlinks, and topical coverage fall short of what gets referenced in AI answers.

Step 4: Track daily and iterate: Monitor how often your site appears in ChatGPT and Google AI Mode for your target prompts. Watch visibility improve as fixes go in.

Most brands chasing AI visibility focus entirely on content. The ones actually showing up fixed their technical foundation first. You can try Semrush free for 7 days.

🍳 A Food AI Startup Claims It Just Had Its ChatGPT Moment

KAIKAKU AI published a research paper this week showing its model can understand flavor, cuisine, and texture purely by analyzing how chefs combine ingredients, no chemistry data required.

The Shift:

What the Model Learned - After cleaning 6,653 ingredient entries down to 1,032 foods, the model identified all five basic tastes and ordered peppers by spiciness without any taste labels.

It Also Mapped Cuisine by Region - The model tagged cuisines by region purely from ingredient patterns, the same way a chef might recognize a dish's origin from its flavor profile alone.

The Three Use Cases - Menu development, recipe innovation, and flavor pairing, work that has traditionally relied entirely on chef intuition rather than data, are the three flagged applications.

The Bigger Play Is Robotics - KAIKAKU is combining this flavor AI with its robotics arm, pitching the combination as autonomous food infrastructure for commercial kitchens.

The claim of a ChatGPT moment is bold but the underlying finding is genuinely interesting, a model learning taste from structure rather than chemistry is a meaningful step forward.

Together with Agentic Marketing Summit

Five days to change how you think about marketing forever.

Not a webinar. Not a panel.

Five days of live sessions with the people actually running thousands of marketing campaigns a month — and handing you the playbook to do it yourself.

Join the Agentic Marketing Summit to be part of the biggest change in marketing to date!

Join Free Today

🔨AI Tools for the Shift

📄 WorkProcedures - Generate audit-ready SOPs in minutes using AI trained on thousands of real industry procedures.

📊 PaperPlot - Create high-quality research diagrams and methodology visuals from text using an AI-driven framework.

🤖 Skymel LEGATO - Turn prompts into production-ready AI agents in hours with a fast, scalable workflow.

🛡️ AI QA Monkey - Run deep security audits to detect hidden vulnerabilities across ports, passwords, and system files instantly.

🛒 Sprite - Automate your ecommerce content operations with AI that creates and manages content daily.

🚀Quick Shifts

🤖 SoftBank plans Roze AI,  a robotics-driven company to automate data center construction, aiming to boost efficiency, scale infrastructure, and potentially reach a $100B IPO amid rising AI and automation demand.

⚖️ Elon Musk faces courtroom scrutiny as his past tweets contradict sworn testimony, weakening his OpenAI lawsuit and exposing inconsistencies around AI ambitions, investments, and shifting positions on profit, control, and safety. 

💼 Microsoft secures long-term, royalty-free access to OpenAI models, enabling Nadella to scale AI aggressively, monetize via Azure, and maintain dominance despite rising competition and OpenAI’s expanding multi-partner strategy. 

📈 Google adds 25M subscriptions in Q1, hitting 350M total as YouTube Premium and Google One drive growth, signaling a major shift from ad-based revenue toward subscription-led monetization and bundled AI services.

🔍 Google Search hits an all-time high queries as AI features like Gemini and AI Overviews boost usage, driving strong revenue growth and signaling a major shift in how users interact with search.

🧩 Prompt of the Day

Dashboard Setup That Makes Reporting Faster and Clearer

Most dashboards become messy data dumps. A strong dashboard shows the exact metrics teams need to make faster decisions.

Turn scattered reports into one clean source of truth.

Paste the prompt: Drop this into ChatGPT, then fill in your business and data details.

Prompt to paste

Create a real-time dashboard setup plan for [Insert brand, campaign, or business]. Define the key metrics to track, such as revenue, traffic, conversion rate, CAC, ROAS, retention, or engagement. List the data sources needed, such as Google Analytics, Shopify, Meta Ads, CRM, or email platforms. Recommend visualization types like scorecards, line charts, funnels, tables, and comparison charts. Focus on making reporting easier, faster, and more useful for decision-making.

🤳AI Nugget of the Day

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:

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