The Cloud Pod | Weekly AI & Cloud News on AWS, Azure & GCP · Justin Brodley, Jonathan Baker, Ryan Lucas and Matt Kohn | Cloud Computing & AI News

AWS CloudWatch Finally Hits Snooze

·1 hr 12 min·4 clips
Sonnet 4.6 is now GA with a $1 million token context window and 94% on insurance benchmarks.
1. The Cloud Pod episode 343 centers on AWS CloudWatch finally learning to hit snooze, recorded February 17th, 2026. 2. Justin, Jonathan, Ryan, and Matthew host the show and matter because they steer the weekly AWS, Azure, GCP, and AI news roundup. 3. The episode asks what changes when cloud utilities, web browsing, and coding are increasingly shaped by AI agents. 4. The first story covers a Notepad markdown vulnerability tied to Microsoft adding lightweight markup support in May. 5. The hosts say the flaw can trigger embedded malicious links when a crafted markdown file opens in Notepad. 6. They also connect the issue to AI-powered text writing and other network-dependent features in Notepad. 7. Chrome's WebMCP story follows, describing a protocol for exposing structured tools and actions to AI agents. 8. The hosts explain declarative and imperative APIs for forms, checkout flows, booking systems, and support tickets. 9. They say the protocol exists to reduce the reliability problems agents face when parsing raw HTML and DOM elements. 10. The discussion then shifts to the business model of the web, with the hosts questioning where ads and monetization go when agents choose the best price directly. 11. Anthropic's numbers dominate the next segment, including a $30 billion Series G, a $380 billion post-money valuation, and $14 billion in run-rate revenue. 12. The hosts note that Anthropic serves eight of the Fortune 10 and more than 500 customers spending over $1 million annually. 13. They also cite Cloud Code at $2.5 billion in run-rate revenue and 4% of all public GitHub commits worldwide. 14. A major theme is enterprise pull, including HIPAA-compliant Cloud for Enterprise and availability across Bedrock, Vertex AI, and Azure Foundry. 15. Sonnet 4.6 is presented as generally available at the same price as Sonnet 4.5, with a $1 million token context window in beta. 16. The hosts highlight 94% on insurance benchmarks, 70% user preference over Sonnet 4.5, and a reported 80.2% SWE Bench Verified score. 17. The tone is conversational and comparative, with the four hosts interrupting each other and using concrete examples from their own work. 18. The format mixes headline reading with short reactions, side comments, and practical examples from PowerPoint, spreadsheets, and coding workflows. 19. People interested in cloud AI tooling and enterprise model releases will get the most from this episode. 20. Listeners wanting tightly edited single-topic analysis may skip the long back-and-forth.

As heard by us

A brisk cloud-news panel that treats convenience, security, and AI hardware like real tradeoffs.

This CloudPod installment moves from cloud-news chatter into a sharp look at a Notepad markdown security issue, then widens into the risk of supply-chain attacks. It also takes on the question of whether an Apple silicon box can really serve as a serious local LLM machine.

Read the full review in PlayNext →

Why you'd press play

Cloud news, a Notepad security scare, and a local-LLM hardware debate, all with a dry grin.

Read the full recommendation in PlayNext →
Listen to the show on