1
2 Comments

Daily AI Quick Links (02/20/2024)

Welcome to today's SMMRY. The key highlights:

  • Reddit's new AI deal to sell user content 🤖

  • AI determines sex from brain scans 🧠

  • OpenAI valued at $80 billion 💰

🚀 Headlines & Launches


OpenAI Valued at Over $80 Billion in Latest Deal (1 minute read)

OpenAI has finalized a deal that places its value at a staggering $80 billion or more. The deal involves the sale of existing shares through a tender offer spearheaded by venture firm Thrive Capital.

Reddit has a new AI training deal to sell user content (2 minute read)

Reddit will let "an unnamed large AI company" have access to its user-generated content platform in a new licensing deal. The deal is worth over $60 million on an annualized basis.

FTC Proposes New Protections to Combat AI Impersonation (3 minute read)

The FTC is finalizing a rule, banning government and impersonation fraud and seeks to extend protections to individuals.

🧠 Research & Innovation


Groq runs Mixtral 8x7B-32k with 500 T/s (Demo)

Groq is serving AI models at insanely fast speeds. The claim is that using their new LPUs is faster than GPUs for serving LLMs.

GPTScript (Github Repo)

GPTScript is a new scripting language to automate interaction with Large Language Models (LLMs).

AI Determines Sex of Person From Brain Scans (5 minute read)

Researchers developed an artificial intelligence model that accurately determines the sex of individuals based on brain scans, with over 90% success.

🛠️🔥 Top Tools


AI Sound Effects by ElevenLabs - Generate sound effects from videos

DomoAI Video to Video 2.0 - Ghibli-style transformation for videos

SoraHub - Explore the newest Sora-generated videos

CodeRev - Code review as interview made easy.

Adsby - Your AI Co-Pilot in Search Ads


These are the main links from today's SMMRY. Read the full edition or subscribe to the SMMRY newsletter here.

on February 20, 2024
  1. 1

    I wonder why reddit is selling stuff rather than building their own.

    1. 1

      My thoughts exactly. Reddit is sitting on a treasure trove of data they can use to build and power their own LLMs. If I had to guess, it's probably because of a lack of budget tbh.