Skip to main content
Latent Space

One Year of MCP — with David Soria Parra and AAIF leads from OpenAI, Goose, Linux Foundation

·
David Soria Parra

Read time

2 min

Topics

Startups, Design & UX, Artificial Intelligence

AI-Generated Summary

Key Takeaways

  • Remote Authentication Evolution: MCP authentication spec required major revision in June after March launch because combining authentication server and resource server into one made enterprise IDP integration (Okta, Auth0) impossible, forcing separation of concerns for real-world deployment.
  • Enterprise Scale Challenges: MCP handles millions of requests at companies like Google and Microsoft, requiring protocol redesign to support horizontal scaling across pods without shared state dependencies like Redis, moving beyond simple single-server architectures that worked initially.
  • Progressive Discovery Pattern: Instead of dumping all tools into context causing bloat, MCP enables models to request information incrementally—give initial data, let model decide what more it needs, similar to how skills explore different functions before connecting to actual data sources.
  • Foundation Governance Model: Agentic AI Foundation uses traditional open source maintainer approach with eight-person core team making decisions, not IETF-style open consensus which takes years, allowing faster iteration while AI technology evolves rapidly compared to three-year OAuth 2.1 standardization processes.

What It Covers

MCP celebrates one year since public launch, with David Soria Parra from Anthropic discussing protocol evolution, enterprise adoption at scale, and the donation to the newly formed Agentic AI Foundation alongside OpenAI and Block.

Key Questions Answered

  • Remote Authentication Evolution: MCP authentication spec required major revision in June after March launch because combining authentication server and resource server into one made enterprise IDP integration (Okta, Auth0) impossible, forcing separation of concerns for real-world deployment.
  • Enterprise Scale Challenges: MCP handles millions of requests at companies like Google and Microsoft, requiring protocol redesign to support horizontal scaling across pods without shared state dependencies like Redis, moving beyond simple single-server architectures that worked initially.
  • Progressive Discovery Pattern: Instead of dumping all tools into context causing bloat, MCP enables models to request information incrementally—give initial data, let model decide what more it needs, similar to how skills explore different functions before connecting to actual data sources.
  • Foundation Governance Model: Agentic AI Foundation uses traditional open source maintainer approach with eight-person core team making decisions, not IETF-style open consensus which takes years, allowing faster iteration while AI technology evolves rapidly compared to three-year OAuth 2.1 standardization processes.

Notable Moment

Anthropic internally uses MCP extensively through a custom gateway where employees deploy their own servers for everything from Slack summaries to biannual survey analysis, with 90% of internal MCP servers unknown to the core team—validating the original vision of self-service tooling.

Know someone who'd find this useful?

Episode Transcript

Everyone, welcome to the Lit in Space podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Lit in Space. Hey. And here we are joined finally in the studio for the first time. Welcome back, David, from Anthropic slash MCP. Yeah. Hey. Well, nice to finally talk to you in person. Like, last time, like, a year ago, it was over VC, and this is wave fun. I watched it back those eight months. It's it's been a crazy eight months, and, I think we just celebrated, like, the one year anniversary of MCP. Yes. We did. At least the public announcement. Yeah. And also last night or yesterday was the agentic AI omniscient launch. Yeah. That was nice. It was a nice event. It was nice to see the Entropic office Yeah. And I've been. You like that? Yeah. It's very good food. I I would say, in terms of my food bench, Entropic does rank over OpenAI. Yeah. So At least that that that's what we have going for us. Awesome, man. Do you wanna give just a quick overview of what's happening with MCP and how you're donating it to the foundation, and then we'll do kinda like a one year recap of the protocol itself, and then we'll have the rest of the leads from the foundation join us to do more of the high level. Yeah. And the yeah. That sounds good. Yeah. I mean, the where where we at at the moment, we have done, like, a year, like, a year ago, we launched it, and then we had this, like, crazy adoption over the last year now, which it felt like an eternity, honestly. But we have this, like, crazy, growth and, like, adoption, you know, through initially, through, like, Thanksgiving and Christmas very early with a lot of, like, builders building MCP, and then, you know, you had, like, the first big clients coming in, like, Cursor and Versus Code. And then, like, you had this, like, inflection point around April with, like, Sam Altman and, Satya and, Sundar and all, posting about, like, MTP and that they're gonna adopt MTP at Microsoft, at Google, at OpenAI, and that was really, like, the big inflection point. So Yeah. But in all of the time, you also had to do a lot of work on the protocol itself, right? We, like, we we moved we launched originally as, like, basically local only, we could, like, build local MCP servers for cloud desktop, but then we like in March, we moved into like, how can you do remote MCP server? So connect like really about like to a remote server and introduced like the first, iteration of authentication. And then in June, we revisited that and, like, improved it quite a little bit so that it works better for, you know, for enterprises in particularly. And we were very, very lucky that in that time from March to, …

Get the full transcript (20,062 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Latent Space transcripts →

Get Latent Space summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

More from Latent Space

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best AI Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's Startups & Product Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into Latent Space.

Every Monday, we deliver AI summaries of the latest episodes from Latent Space and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime