Guru99 AI Report โ€บ News Letter โ€บ Current Edition

Big Tech Spends $130B on AI in Q1 as Cloud Demand Surges

ALSO: Claude Security launches, Gemini ships file gen

Guru99 AI Report

Welcome to Guru99 AI Report!

Top Story: This week’s AI updates go beyond product launches. From cloud constraints and security risks to smarter workplace agents and changing job anxieties, here’s what these shifts could mean for the tools you use every day.

๐Ÿš€ Big Tech’s $130B AI Boom Is Just Getting Wild

Big Tech AI capex boom
Brief Buzz: On April 29, 2026, Microsoft, Google, Meta, and Amazon dropped Q1 earnings on the same day โ€” and the message was identical: AI demand is outpacing what the world’s richest companies can physically build. Combined, they spent roughly $130 billion on infrastructure in just three months โ€” matching NVIDIA’s entire 2025 annual revenue.
๐Ÿ’ก Why Should You Care?
Cloud prices likely won’t drop anytime soon โ€” hyperscalers can’t keep up with demand, so expect zero enterprise discounting. The flip side: multi-model AI is now default (Claude + GPT auto-routing inside Microsoft Copilot), meaning smarter, more reliable apps. But if AI usage growth slows, this could become tech’s largest infrastructure overbuild ever.

๐Ÿ”Ž Anthropic Launches Claude Security to Fight AI-Driven Risks

Claude Security launch
Brief Buzz: Anthropic has launched Claude Security in public beta for Enterprise users โ€” a dedicated tool to hunt down code vulnerabilities. Powered by Opus 4.7, it’s a more accessible cousin to the highly restricted Claude Mythos, arriving as AI models grow dangerously capable of both finding and exploiting software flaws.
  • Reasons about code like a security researcher, not just pattern-matching against known threats
  • Offers scheduled scans, audit integrations, and triaged findings โ€” no API setup needed if you already use Claude
  • Integrates with major security platforms including CrowdStrike, Microsoft Security, Palo Alto Networks, SentinelOne, and Wiz
  • Arrives as the White House reportedly opposes Anthropic’s plan to expand Mythos access to 120 organizations over security concerns
  • Rival OpenAI’s GPT-5.5-Cyber is matching pace, currently restricted to “critical cyber defenders”
๐Ÿ’ก Why Should You Care?
The same AI that can patch security holes can also blow them wide open. Every app, bank, and hospital you rely on runs on code โ€” and these tools could either harden that code against attackers or hand cybercriminals a master key. The fact that even the U.S. government is nervous about wider release tells you everything: this is genuinely powerful tech with real-world stakes for everyone online.

๐Ÿค AI Can Do More โ€” But Humans Still Win

Starbucks reverses automation push
Brief Buzz: Starbucks just reversed its automation push, betting baristas and ceramic mugs beat machines. University of Chicago economist Alex Imas argues in a must-read essay that this is the canary in the AI coal mine: when AI makes everything cheap, what humans uniquely offer becomes the new luxury.
  • Starbucks CEO Brian Niccol is hiring more baristas and pulling automation from busy cafรฉs, telling The Guardian that handwritten notes and great seats drove customer satisfaction
  • In Imas’s exclusivity experiments, people paid roughly 2x for identical items when others were excluded โ€” exclusivity is the new premium
  • AI-generated art earned only a 21% exclusivity premium versus 44% for human-made art
  • A “relational sector” is emerging โ€” teachers, nurses, therapists, craft brewers, live performers โ€” where the human IS the product
  • The catch? Distribution may mirror Spotify, where 80 top artists earn $10M+ yearly while the 100,000th-ranked artist earned just $7,300 in 2025
๐Ÿ’ก Why Should You Care?
As AI commoditizes routine work, human qualities โ€” relationships, craft, presence โ€” become the premium. Your career may hinge less on outpacing AI than on what only you can offer. But without thoughtful policy, this relational economy risks becoming another winner-take-all platform game.

๐Ÿš€ Perplexity Brings Its AI Agent to Teams

Perplexity Personal Computer
Brief Buzz: At its Ask NYC event, Perplexity rolled out major upgrades to Personal Computer, its AI agent that works across local files and native apps โ€” not just the cloud. CEO Aravind Srinivas reframed it from Steve Jobs’s “bicycle for the mind” into something closer to three Ferraris.
  • Personal Computer is now open to Pro subscribers on Mac (previously Max-tier only), plus a Microsoft Teams integration and a beta Excel side panel
  • A new 1Password partnership lets the agent act inside password-protected tools without exposing credentials to the AI model
  • 50+ prebuilt Workflow templates can be shared, scheduled, and run asynchronously, alongside fresh Databricks and Snowflake connectors
  • A dedicated Computer for Financial Services tier brings your own licenses for Carbon Arc, Daloopa, Morningstar, and PitchBook, plus an Equity Research Council workflow that mimics a panel of analysts
  • Perplexity says its agent has already performed $2.8B in labor-equivalent work since launch
๐Ÿ’ก Why Should You Care?
Agentic AI is graduating from chatbot novelty to genuine workplace muscle. With Computer now reaching more users โ€” and safely plugging into the apps, passwords, and licensed data people actually use at work โ€” routine research and ops tasks could shrink from days to hours. The real shift: enterprises finally have an AI agent they can trust with the keys.

๐Ÿ“Š Why AI’s Top Time-Savers Are the Most Worried

AI worker anxiety study
Brief Buzz: A new Anthropic survey of 80,000+ workers reveals a surprising twist: the employees gaining the most from AI tools like Claude are also the ones most anxious about being replaced by them. Early-career workers and engineers top the worry list.
  • The study links Anthropic’s Economic Index usage data with responses from 80,508 workers on how AI is reshaping their roles.
  • Workers whose jobs rely heavily on Claude reported displacement fears 3x higher than light users, with engineers leading the anxiety charge.
  • Most respondents said productivity gains bring faster tasks and free time, but also expanded scope and more overall work.
  • Early-career workers voiced the loudest concerns, echoing signs of a hiring slowdown for recent U.S. graduates.
๐Ÿ’ก Why Should You Care?
This flips the usual narrative on its head. The people closest to AI โ€” and benefiting most from it โ€” are the most worried about their future. If you’re early in your career or in a tech-heavy role, this hints at a tougher job market ahead, with no clear fix in sight.

๐Ÿš€ Gemini Now Builds Docs, PDFs & Sheets in Seconds

Gemini file generation
Brief Buzz: Google has handed Gemini a long-requested superpower: the AI app can now generate fully-formatted, downloadable files directly from a single prompt. CEO Sundar Pichai flagged the rollout on X, pitching it as a way to ditch the copy-paste-reformat dance for good.
  • Supports PDFs, Microsoft Word, Excel, Google Docs, Sheets, Slides, plus CSV, TXT, RTF, Markdown, and LaTeX
  • Files download directly to your device or export straight to Google Drive
  • Available globally to every Gemini user, including individual Workspace accounts
  • LaTeX support is a quiet win for STEM students and researchers drafting scientific papers
  • Move comes as Google races Claude and ChatGPT on AI productivity features and file-export capabilities
Why Should You Care?
Whether you’re drafting a budget, building a study guide, or polishing a client proposal, Gemini just became a true one-stop workshop โ€” saving real time for students, professionals, and anyone who basically lives inside documents.
Krishna Rungta
Facebook LinkedIn Twitter/X

Hey! I’m Krishna Rungta

Founder of Guru99.com, Editor-in-chief & Technology Expert

Was this email forwarded to you? Sign up for free here.

Summarize this post with: