Microsoft embraces its rival's protocol

Plus: Google’s new whitepaper on agent design, how to tell an agent from a sham, and more

Welcome back to Building AI Agents, your biweekly guide to everything new in the field of agentic AI!

“Sure AI can now do this, but there’s no way it will ever be able to do that.” Spoiler alert: it will be doing that in 6 months.

In today’s issue…

  • Microsoft yields to Google in the agent protocol war

  • Google’s new whitepaper on agent design

  • A taxonomy of agent failure modes

  • How to tell an agent from a sham

…and more

📅 AGENT OF THE WEEK

🧠 Lead Researcher Agent

Problem: A new meeting gets booked—now you're spending 30+ minutes Googling names, checking LinkedIn, and digging for company info, just to avoid walking in without context.

Solution: The Lead Researcher Agent handles the prep for you. It:

  • Detects every new meeting as it’s booked

  • Researches the person and their company

  • Scores lead quality at a glance

  • Emails you a crisp prep brief before the call

Result: ⏱️ Save 30+ minutes per meeting. Show up prepared—with the who, what, and why already in hand.

As a subscriber, you’re invited to a free 30-minute strategy call with Orynt AI—to see how agents like this could work for you. 

Subscribe to keep reading

This content is completely free, but you must be subscribed to Building AI Agents to continue reading.

I consent to receive newsletters via email. Terms of use and Privacy policy.

Already a subscriber?Sign in.Not now