Urgent.News

What's breaking now, across thousands of outlets.

AI

LinkedIn cofounder says AI infrastructure is the 'only reason we're not in a recession'

Reid Hoffman says he's thankful for the data center buildout in the US, while many Americans are souring on it.

LinkedIn co-founder Reid Hoffman expressed gratitude for the extensive data center construction occurring across the United States. Hoffman attributed the absence of a recession to this infrastructure development. While some Americans are critical of the data center buildout, Hoffman remains thankful for the capital behind it. He stated during an appearance on the Newcomer podcast, "The only reason we're not in a recession is because of the capital behind the AI infrastructure buildout."

Most of this capital does not directly benefit the areas where data centers are being built, but rather is distributed more broadly throughout the country, creating jobs for various professionals in the construction sector. Despite the recent negative perception of data centers among rural Americans and some political pushback, Hoffman emphasizes the importance of fair treatment for local communities.

He believes that data center builders must work in good faith with these communities to ensure mutually beneficial agreements. Even though the US economy faced a challenging September jobs report, unemployment remains low, and overall economic growth is still robust. Real gross domestic product grew by 2.2% in the second quarter, and personal income is also increasing.

Written by urgent.news from Business Insider's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at businessinsider.com →

More in AI

Can JEV / TEV replace Embedding for intent recognition? | Jev / TEV 能不能干掉 Embedding 做意图识别?

Jev / TEV 能不能干掉 Embedding 做意图识别? 不能完全替代,但可以干掉一大半传统 Embedding 意图识别的场景,二者有明确分工,不是简单谁取代谁。 先把两套方案本质讲通俗: 1、老方案:Embedding 向量做意图识别 流程:句子 → 向量化 → 向量相似度检索,匹配预先写好的意图库 原理: 语义相似度匹配 ,靠向量空间远近判断属于哪个意图 适合:意图集合…

Turning Claude Code into a Writing Harness with Claude Mods and an Output Style

Writing articles in Claude Code has advantages that chat on claude.ai doesn't. You can build your own harness of skills and rules, and you have a lot of freedom over what context Claude gets.

  • Claude Code offers advantages for article writing with custom harnesses.
  • Mods control Claude's input, output style defines conversational tone.
  • zenn-writing profile reduces skill listing and agent list size.

How AI Learns: Forward Pass and Loss Explained with 2 + 1

Our inputs are 2 and 1, and the correct answer is 3. But the network initially predicts 0.7. Why? A neural network doesn’t automatically know the rules of addition.

  • Forward pass calculates prediction using inputs and network weights
  • Loss measures discrepancy between prediction and correct answer
  • Backpropagation adjusts weights to minimize loss

More from Saturday 3 October →