How much revenue does OpenAI generate?

Good Morning AI Runners

Here's what we've got for you today:

• OpenAI Eyeing $29B Valuation

• Legal NLP Dataset (39k+ examples)

OpenAI Eyeing a Valuation of $29 Billion (currently at $14 Billion)

According to sources familiar with the matter, OpenAI is in discussions to sell its existing shares through a tender offer that would value the company at approximately $29 billion.

This would more than double the value of a previous tender offer in 2021 that valued the company at around $14 billion!

The tender offer is expected to total at least $300 million and will involve the purchase of shares from existing shareholders such as employees by two VC firms Thrive Capital (Joshua Kushner) and Founders Fund (Peter Thiel).

If the deal goes through, OpenAI will become one of the most valuable US startups on paper, despite generating little revenue.

In fact, it is estimated to become the sixth most valuable unicorn in the US:

How much revenue does OpenAI generate?

While OpenAI is a private company and little information is available about its current revenue, Reuters has reported that the organization has projected revenues of $200 million in the coming year and $1 billion by 2024. These projections are based on information from sources who were briefed on OpenAI's recent pitch to investors. It is worth noting that these projections have not been confirmed by OpenAI.

P.s. $29 Billion would make for a 145X multiple and a very high valuation, but did you know Roblox was once valued at $42 Billion?

Legal NLP Dataset With Over 39,000 Examples

Robo Lawyers anybody?

Legal datasets are extremely expensive because lawyers are, and this has bottlenecked legal NLP.

To address this, the Merger Agreement Understand Dataset (MAUD), was recently released with over 39,000 multiple-choice reading comprehension examples for 152 merger agreements that have been manually labeled by legal experts.

The dataset was created with the help of the American Bar Association; without their help the dataset would have cost over $5,000,000 to create.

MAUD has substantial room for improvement and can could serve as a research challenge for NLP researchers without any legal background.

Read more in the paper linked below 👇

Dataset and Baselines: https://github.com/TheAtticusProject/maud/

Paper: https://arxiv.org/abs/2301.00876

Pic of the day:

That's it from RuntheAI for today.

THANK YOU FOR READING AND SEE YOU TOMORROW, SUBSCRIBE TO STAY UPDATED!