Korva is a scroll of the internet’s best ideas.

One Person's Department

The company serves lots of customers and needs very traditional marketing work: sales enablement and content. The person in charge was interested in creating leverage for himself, and that led him to build it.

Companies Moving Faster

Founders across Conviction's portfolio report the same thing: the company is just going way faster. It looks like the transformation that already happened in software engineering, and the same pattern is expected to reach every other function as agents and products take over more of the mundane work.

Where Saved Time Goes

Jevons paradox points to where the saved time goes: it gets repurposed into more work. Sarah Guo, who invests in the company, works more since AI made her more productive. Jensen's view is that everyone will end up more employed, provided people get access and education to the tooling that makes it possible.

Why does efficiency lead to more work?

In 1865, the economist William Stanley Jevons noted that more efficient steam engines had increased Britain's coal use rather than reducing it. Cheaper use made new applications worthwhile, and total demand rose. The same pattern, now called the Jevons paradox, has been described for lighting, computing, and highway lanes.

How does an AI agent do marketing work?

An agent is a language model placed in a loop where it can plan a step, use a tool, and read the result. For marketing, the tools can include a document editor, a customer database, a search engine, and an email system. The model drafts a piece, checks it against product notes, and revises it. A person sets the goals and reviews the output at key points.

Why didn't ATMs end bank teller jobs?

Economist James Bessen found that the number of US bank tellers grew modestly between 1980 and 2010 despite the spread of ATMs. Each branch needed fewer tellers, but cheaper branches meant banks opened more of them. The tellers' work shifted toward sales and customer service. ATMs were first widely deployed in the 1970s.

More from Invest Like the Best with Patrick O'Shaughnessy

Tech & AIPower is waiting on public buy-inPower for AI data centers is held up by regulation and public buy-in, since Americans and entrepreneurs have built new technology very quickly before.
Enjoyed it?