nafyi.About
Topics / AI infrastructure

A TOPIC, IN CONTEXT

AI infrastructure

Compute, model scaling and the economics of AI systems.

3 people · 2 sources · 3 viewpoints

Viewpoint map

Explore by person. Select two or three to compare.

3 people · 2 sources · 3 viewpoints

Nick Kuhn

MCP gateways control and scale tool access

An MCP gateway provides a way to easily control and scale access to MCP servers within a cloud platform. Organizations may run dozens of MCP servers that get registered and bound to specific gateways for centralized management.

From AGENTS.md to Enterprise Deployment with Nick Kuhn

Supporting evidence

Original excerpt

if you think about an MCP gateway as a way to easily control and scale access to MCP servers, especially within you're in a cloud, I guess, platform. So we may have, like, 30 or 40 MCP servers that we run

Nick Kuhn·

From AGENTS.md to Enterprise Deployment with Nick Kuhn
Context

Open original · 23:49 (opens in a new tab)Transcript source (opens in a new tab)

Start time comes from the supplied transcript. Playback alignment is awaiting review.

Open the episode and seek to 23:49.

Share insightCheck this claim

Dario Amodei

Linear scaling of three ingredients

Amodei compares model scaling to a chemical reaction requiring three ingredients—bigger networks, longer training times, and more data—to be scaled linearly together. Scaling only one ingredient causes the process to stall because the other reagents run out.

Dario Amodei, Amanda Askell & Chris Olah on AI Development and Safety

Supporting evidence

Original excerpt

almost like a chemical reaction, you have three ingredients in the chemical reaction and you need to linearly scale up the three ingredients. If you scale up one, not the others, you run out of the other reagents and the reaction stops.

Dario Amodei·

Dario Amodei, Amanda Askell & Chris Olah on AI Development and Safety
Context

Open original · 7:17 (opens in a new tab)Transcript source (opens in a new tab)

Start time comes from the supplied transcript. Playback alignment is awaiting review.

Share insightCheck this claim

Chris Olah

Neural networks are grown, not programmed

Neural networks are not directly programmed but grown. Researchers design architectures as scaffolds and set loss objectives as directional targets, but the resulting system is more like a biological organism whose internal workings must be studied rather than code that was written.

Dario Amodei, Amanda Askell & Chris Olah on AI Development and Safety

Supporting evidence

Original excerpt

we don’t program, we don’t make them, we grow them. We have these neural network architectures that we design and we have these loss objectives that we create. And the neural network architecture, it’s kind of like a scaffold that the circuits grow on.

Chris Olah·

Dario Amodei, Amanda Askell & Chris Olah on AI Development and Safety
Context

Open original · 4:18:02 (opens in a new tab)Transcript source (opens in a new tab)

Start time comes from the supplied transcript. Playback alignment is awaiting review.

Share insightCheck this claim

These are individual perspectives, not a measure of consensus. Source material stays in its original language.

Original conversations

CONVERSATION

From AGENTS.md to Enterprise Deployment with Nick Kuhn

Practical AI

VIDEO INTERVIEW

Dario Amodei, Amanda Askell & Chris Olah on AI Development and Safety

Lex Fridman Podcast