Kailash Nadh about MCP
Kailash Nadh, Zerodha's CTA, has written an interesting blog post about MCP where he presents different scenarios of how MCP can be used, and also talks about the rapid adoption.
The funny thing is, as a technical construct, there is nothing special about MCP. It is a trivial API spec which has in fact suffered from poor design and fundamental technical and security issues from the get go. It does not matter if its internals change, or it even outright gets replaced by some XYZ tomorrow. Questions about privacy, security, correctness, and failures will continue to loom for a good while, irrespective of whether it is technically MCP or XYZ.
He talks about how, traditionally, connecting different software systems required extensive manual coding but MCP allows connecting services instantly.
I liked that he also talked about the concerns, as he worries about:
- AI systems making real-world decisions with minimal human oversight
- Questions of accountability when things go wrong
- Privacy and security implications
One might imaginatively call it … SkyNet.
He also playfully compares MCP to SkyNet while calling it a "global, interconnected, self-organising meta system".
Overall, it's a balanced post sharing his technical excitements with genuine concerns about such AI systems getting full access to real-world services and decision-making power.
By the way, I almost forgot to mention that Zerodha itself has launched Kite MCP a few days ago.
- ← Previous
Remix is moving on from React - Next →
Collecting funny memes
Comment via email