💡 Alphabet Just Bet $85B That You're Still Thinking About AI Wrong

In partnership with

☕ Morning! 💡 Your Weekly 5-Minutes of Caffeine and Tech Clarity

Featured Event 📊 

Tokenomics for the Deeply Skeptical

AI experimentation made sense when the costs were low.

At scale, that same approach gets expensive fast.

Save your spot and figure out what engineering leaders are doing today.

(Before spots run out [Click here])

The engineering leaders managing this well made deliberate decisions about where AI investment earns its keep.

Find out what works on June 15.

Quick Hits 🎯

• 🔋 Energy strategy is now a boardroom conversation. AI's power appetite is forcing companies to plan infrastructure the way they plan headcount.

• 💰 Alphabet closes an $85B raise for its AI business, the largest capital commitment in the company's history.

• 🤖 Anthropic reports 80% of its production code is now written by Claude. That number is moving faster than most engineering orgs are ready for.

• ⚖️ Courts are drowning in AI-generated legal filings - the legal system is building process infrastructure in real time.

• 💻 Google's Gemma 4 12B runs multimodal tasks entirely on a standard 16GB laptop. No cloud required.

• 🔐 Anthropic drops an open-source framework for AI-powered vulnerability discovery. Security tooling just got a community layer.

• ✂️ GitLab cuts 14% of its workforce while doubling down on AI platform capabilities. This is what "right-sizing for AI" looks like

• 🤝 Lovable signs a multiyear deal with Google Cloud to scale usage 5x. Platform bets are getting made fast

🎁 + 2 other stories you might find useful

Our Partner 🎉 

Stop making AI decisions in the dark.

Leadership is asking: are we getting value from AI? Which tools are worth the spend? Where are we exposed? Right now, most teams have no idea.

Harmonic Security Usage Explorer changes that.

You get a complete picture of how your organization uses AI, automatically categorized into custom tasks and use cases.

You’ll see the projects being worked on, who’s using what tools, where AI investments are driving value, and where employees are engaging in risky behavior.

CIOs can rationalize spending and cut wasted licenses. CISOs can pinpoint where risk exists and neutralize it. AI committees can show exactly how their efforts are paying off.

See it in action

The Big Picture 🖼️

💡 AI Has an Energy Problem. And It's Now a Strategy Problem.

As workloads scale, the cost and reliability of the power behind those models stops being an ops concern and starts being a business risk. Teams that map energy provisioning to product roadmaps now will have a structural advantage over those who treat it as a vendor problem.

The companies that figure this out first won't just have lower bills. They'll have a narrative customers and partners can trust. The framing most executives are still missing is that energy strategy isn't about sustainability optics. It's about AI reliability at scale.

The takeaway: if your infrastructure roadmap doesn't include energy provisioning, your AI roadmap isn't finished.

💡 $85 Billion Is a Signal, Not a Number.

Alphabet's raise for its AI business isn't just about buying more GPUs. It's about funding the full stack. Things like proprietary models, smarter inference, deeper enterprise integration, can close the gap on Google Cloud before competitors.

When capital moves at this scale, it tells you something about where enterprise AI budgets are actually going. The strategic read most analysts are skipping is that this isn't Alphabet going on offense. It's Alphabet building a moat.

The takeaway: capital is concentrating around platforms that can prove efficiency and integration together - not just raw model performance.

💡 80% AI-Authored Code Is a Governance Wake-Up Call.

Anthropic saying 80% of its own production code is now written by Claude isn't a brag - it's a data point that most engineering organizations aren't ready for.

The speed is achievable. But speed without governance creates technical debt you can't audit and risk you can't explain to your board. What separates the teams already running at this velocity is that they built the guardrails before they scaled the acceleration.

The takeaway: CI/CD pipelines built for human-authored code will break under AI-authored volume. The orgs adjusting now have a six-month head start.

💡 The Legal System Is Building Its Own AI Infrastructure.

Courts are being flooded with AI-generated filings, and the response isn't resistance - it's process design. Judges and clerks are building scalable adjudication systems to handle volume that would have been impossible five years ago.

For engineering and product teams, this is the clearest signal yet that governance as a product feature is no longer optional. The companies that can document, audit, and explain AI outputs will spend less time in discovery and more time shipping.

The takeaway: compliance tooling isn't overhead anymore. It's competitive positioning.

💡 On-Device AI Just Changed the Enterprise Procurement Conversation.

Google's Gemma 4 12B analyzing audio and video entirely on a 16GB laptop isn't a benchmark flex. It's a proof point that enterprise AI deployment no longer requires a cloud contract for every workflow.

For industries with data sovereignty constraints, latency requirements, or tight procurement cycles, this is the architecture shift worth watching. The question isn't cloud vs. edge anymore, it's which workloads belong where.

The takeaway: if your AI strategy assumes cloud for everything, it's time to revisit the map.

💡 GitLab's Layoffs Are a Blueprint, Not Just a Headline.

Cutting 14% of staff while accelerating platform investment for AI workloads is a specific kind of organizational bet. GitLab isn't shrinking, it's restructuring around what scales and what doesn't.

This is the operational model emerging across high-growth engineering orgs: tighter teams, AI-augmented output, platform reliability as the north star. The headcount math changes when AI doubles individual throughput.

The takeaway: org design built for pre-AI productivity assumptions is already obsolete.

💡 Open-Source Security Tooling Is Becoming a Collective Defense Play.

Anthropic releasing an open-source framework for AI-powered vulnerability discovery isn't just a goodwill move. It's a signal that the security layer in AI systems needs community hardening to keep pace with the threat surface.

Teams that participate in open tooling standards and contribute back will have security postures that proprietary-only shops can't match. The harness is already live, and the teams using it now will shape how enterprise AI security gets defined.

The takeaway: security at the AI layer is still an open problem. And the open-source community is where the answers are getting built.

💡 Cloud Partnerships Are Becoming Moats, Not Just Cost Lines.

Lovable's multiyear deal with Google Cloud to scale usage 5x isn't a vendor relationship - it's a strategic lock-in that compresses the gap between AI product ambition and infrastructure reality. A 146-person company crossing $400M ARR, built almost entirely on cloud-native AI.

The details of how this deal is structured, like the marketplace distribution, Wiz security integration, dual model access - explain why cloud-native AI companies are growing faster than their infrastructure-agnostic peers.

The takeaway: your cloud strategy is now part of your product strategy, whether you've decided that or not.

Trending Tools 📈

aquasecurity/trivy (+255 ⭐ this week, 🔷 Go) Link

• Comprehensive vulnerability scanner for containers, Kubernetes, code, and more - with SBOM and secret detection built in

• Helps teams ship faster with confidence: scan your entire software supply chain before anything reaches production

n8n-io/n8n (+181 ⭐ this week, 🔷 TypeScript) Link

• Fair-code workflow automation with AI capabilities, self-hosted or cloud, 400+ integrations

• The alternative to Zapier for teams that want logic, not just triggers - increasingly used to automate AI agent pipelines

freeCodeCamp/freeCodeCamp (+115 ⭐ this week, 🔷 TypeScript) Link

• Open-source curriculum covering math, programming, and CS at no cost

• With AI making code more accessible, teams are using this to upskill non-engineers faster than any internal training program

derailed/k9s (+127 ⭐ this week, 🔷 Go) Link

• Keyboard-driven Kubernetes CLI that makes cluster management feel like a terminal game, not a support ticket

• If your team manages more than two clusters and still uses kubectl for everything, this is the upgrade

caddyserver/caddy (+150 ⭐ this week, 🔷 Go) Link

• Fast, extensible HTTP/1-2-3 server with automatic HTTPS and sensible defaults out of the box

• Teams are switching from Nginx for new deployments because setup that used to take a day now takes an hour

makeplane/plane (+83 ⭐ this week, 🔷 TypeScript) Link

• Open-source, self-hosted alternative to Jira, Linear, and Monday - tasks, sprints, docs, and triage in one place

• The "Jira replacement" search is real - this is the most-watched contender this week

traefik/traefik (+64 ⭐ this week, 🔷 Go) Link

• Cloud-native application proxy with automated TLS, dynamic routing, and service discovery

• A default choice for teams building microservice architectures who want routing that configures itself

Tech Trend of The Week 📊 

🔍 "Anthropic IPO" spiked this week following the confidential SEC filing

On June 1st, Anthropic quietly filed confidentially for a US IPO and search volume jumped as investors, engineers, and enterprise buyers all scrambled to understand what that means.

The filing puts Anthropic's valuation approaching $1 trillion, positioning it alongside OpenAI as a potential benchmark for pure-play AI model companies in public markets.

The signal: when an AI safety company that powers enterprise code generation files for a public offering, it stops being a research story and becomes a balance sheet story.

Enterprise buyers are already using the moment to renegotiate pricing and lock in data governance terms before public market pressure changes Anthropic's incentive structure. The window for favorable contracts is narrowing.

Our Partner 🎉 

Want to get the most out of ChatGPT?

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide.

Hit Reply And Tell Me 💬 

What's the most impressive tech you've seen recently that actually works?

I read every single reply

Go where your customers actually are 📍

Most of the internet copies Reddit.

Blogs, SEO pages, AI models, product reviews. They all pull from the same well.

Reddit is the 3rd biggest site on earth. And it trains the bots your users ask for help.

Odd Angles Media runs Reddit campaigns for over 45 brands each month.

Luckily, we’ve convinced them to give away all of their strategies in a (free) ebook ⬇️⬇️

Grab The Free Ebook (Free) →

Reach the People Who Sign the Checks 💰️ 

40,000+ CTOs and engineering leaders. 96% US-based. 80%+ corporate emails.

They evaluate vendors. They shortlist solutions. They buy.

Advertise with us on Paved → (discount applied)