Multi-Agent vs. Multi-User Challenges

Major SaaS companies including Salesforce, Intuit, and ServiceNow saw stock drops of 45-50% as enterprises shift from bloated software suites to personalized AI agents that users can control directly. Microsoft just capitulated this week, opening Copilot to allow Claude Cowork-style functionality — a clear signal that the "build vs. buy" calculus for enterprise software has fundamentally changed.

Matt Marshall and Sam Witteveen break down why personalization is no longer optional for enterprise products. Companies like Zoom now offer personalized workflows that access your conversation history and profile context. Infrastructure decisions are moving fast: token budgets must account for per-user context, identity management has become the biggest technical challenge for agent deployments, and "skills" (not just MCP) are emerging as the key abstraction layer.

Zoom's Li Juan explains how their AI Companion moved beyond generic templates to user-controlled personalization: tracking opinion divergence in meetings, generating follow-up emails with specific context controls, and giving users explicit prompt examples instead of "good luck with your prompt." This is the new standard. If your product can't reason over which tools to use, which skills to apply, and which context to pull — all personalized to the individual user — you're competing with something that can be built in 10 days (Cowork's timeline).

The agents-are-taking-over reality is here: multi-user agent architectures require thinking about context contamination, security postures for computer-use capabilities, and whether you're building internal agents or buying SaaS that will adapt. Sam's take: "AGI is agentic, and we're well along that continuum now."

🎙️ HOSTS: Matt Marshall | CEO, VentureBeat & Sam Witteveen | VentureBeat

📺 CHAPTERS:

00:00 Intro — The SaaS Apocalypse

00:01:00 The Personalization Imperative

00:02:00 Microsoft Copilot Capitulates to Cowork

00:03:00 From Template Selection to Skill Generation

00:04:00 The Land Grab for User Context

00:05:00 Zoom's Li Juan on Personalized Meeting Intelligence

00:06:00 Why Context = Magic in Enterprise AI

00:07:00 Product-Market Fit in the Agent Era

00:08:00 Metrics That Matter: JP Morgan's 30,000 Agents

00:09:00 Build vs. Buy: The New Calculus

00:10:00 Why Slack Might Win on Agent Identity Management

00:11:00 Zoom's AI Companion: Control Over Randomness

00:13:00 Li Juan on Purposeful Prompts and Reference Control

00:15:00 Multi-Agent vs. Multi-User: The Critical Distinction

00:16:00 LinkedIn's GPU Optimization Strategy

00:17:00 AGI Is Agentic: Where

Presented by Outshift by Cisco Outshift is Cisco’s emerging tech incubation engine and driver of Agentic AI, quantum, and next-gen infrastructure. Learn more at outshift.cisco.com.

About VentureBeat: VentureBeat equips enterprise technology leaders with the clearest, expert guidance on AI – and on the data and security foundations that turn it into working reality.

🔗 CONNECT WITH US

Subscribe to our Newsletters for technical breakdowns: https://venturebeat.com/newsletters

Visit VentureBeat: Venturebeat.com

.

.

.

Subscribe to VentureBeat: 

   /  @VentureBeat  

.

.

Subscribe to the full podcast here:

Apple: https://podcasts.apple.com/us/podcast/venturebeat/id1839285239

Spotify: https://open.spotify.com/show/4Zti73yb4hmiTNa7pEYls4

YouTube: https://www.youtube.com/VentureBeat

Learn more about your ad choices. Visit megaphone.fm/adchoices