The Insane Concept of Scraping the Entire Internet | Jensen Huang ??

The Insane Concept of Scraping the Entire Internet | Jensen Huang ??

Introduction

A lot of people have approached me with a common concern. They say, "Jensen, if you reduce the cost of computing by a million times, then people will buy less of it." But, quite the contrary, what we've observed is that reducing the marginal cost of computing can actually lead to revolutionary advancements and increased utilization of technology.

Reducing Marginal Cost of Computing

The idea is simple yet transformative. If we can reduce the marginal cost of computing down to practically zero, it opens a world of possibilities. Imagine having the computational capability to do something extraordinarily ambitious—like creating large language models designed to extract all digital human knowledge from the internet.

Large Language Models and Internet Scraping

The concept revolves around leveraging large language models to scrape the entire internet for information. This information would then be fed into a computer which could analyze and figure out the underlying knowledge. It’s an idea that sounds almost insane due to its vast scope and ambition.

Transformative Potential

Scraping the entire internet and putting the data into one computer to let it determine patterns and insights on its own is not something anyone would have ever considered unless the marginal cost of computing was negligible. If we can achieve this, the possibilities for artificial intelligence and machine learning could be limitless.

Keywords

• Marginal cost of computing
• Large language models
• Digital human knowledge
• Internet scraping
• Artificial Intelligence
• Machine Learning

FAQ

Q: What is the primary argument made by Jensen Huang in the article?
A: Jensen argues that reducing the marginal cost of computing can lead to groundbreaking innovations rather than reducing the utilization of technology.

Q: Why would reducing the cost of computing open new possibilities?
A: Lower costs would make it feasible to perform huge tasks, like scraping the entire internet to extract and analyze digital human knowledge.

Q: What are large language models in this context?
A: These are sophisticated AI models designed to process and understand large amounts of textual data, potentially uncovering patterns and knowledge from the entirety of the internet.

Q: What transformative potential does this concept hold?
A: If achieved, it could revolutionize artificial intelligence and machine learning, offering unprecedented insights and capabilities.

Q: Is this concept widely considered feasible?
A: Many might initially find it insane due to its ambitious scope, but the feasibility increases significantly with reduced computational costs.