In this blog, EMA’s Herb Blecher shares a hands-on journey from ChatGPT to OpenClaw to reveal a bigger issue: the “agentic enterprise” isn’t a tool problem, it’s an architecture problem.
While AI tools excel at reasoning with the right context, they still struggle with execution, especially across systems. Add in challenges around observability, scalability, and cost, and it’s clear most organizations aren’t ready to move from pilot to production.
The takeaway? Success with agentic AI isn’t about picking the right tool, it’s about building the infrastructure to make them work together.
Other Resources:
See All Resources
The Lost RSAC 2026 Conversations: Business Enablement vs. Security Risk
This blog from EMA's Ken Buckler explores how security measures—past and present—can unintentionally...
Why Cloudflare’s PQC Roadmap is a Wake-Up Call
If you thought "Q-Day" was a problem for the 2040s, it’s time to check your calendar: the goalposts ...
From ChatGPT to OpenClaw: Field Testing the Agentic Enterprise Problem
In this blog, EMA’s Herb Blecher shares a hands-on journey from ChatGPT to OpenCla to reveal a bigge...
AI Was Never Really Coming For Your Job – But Your Job Will Be Different
In this new blog, EMA’s Chris Steffen explains that just like virtualization, cloud, and DevOps befo...
Q-Day Just Got Way More Real
The timeline for "Q-Day"—the hypothetical day when quantum computers become powerful enough to break...
Many Network Engineers Want DDI Vendors to Add Agentic AI Features
In conversations with network teams, one theme keeps emerging: AI-powered management tools are no lo...
Key Takeaways from RSAC 2026
Four themes kept coming up again and again at this year's RSA Conference, and if you're serious abou...
Data Trust: Architecture vs. Policy
Organizations that struggle with trust don't always lack tools; they have a gap between what their p...
A Path Toward Trusting Agentic NetOps
Explore how AI is reshaping network operations and why trust remains the ultimate barrier to adoptio...