Sunday Signal #8

Bing Challenges Google (And Spills Secrets), Opera Enters the AI Space and Mindblowing Video-To-Video

Each week I consume 500+ pieces of content in tech and AI.ย I filter them and send you the best, every Sunday.

Hey friends ๐Ÿ‘‹ ,

If there's one thing I love more than the latest generative search AI being released each week, it's writing these emails. I hope you're all having a smashing weekend. It's been a busy weekโ€”let's get to it.

AI Predictions for 2023, 3D Models and Small Delights

DALLยทE prompt: "draw a black and white sketch of the best newsletter to learn about artificial intelligence"

Noteworthy News

  • ๐Ÿ”— Link: Can Bing and OpenAI challenge Google? Microsoft's Satya Nadella weighs in with the WSJ on how AI is going to change search.

  • ๐Ÿ”— Link: Speaking of which, Kevin Liu got Bing (Sydney?!) to spill its 'secret' prompt.

  • ๐Ÿ”— Link: Opera enters the generative AI space with new features in browsers and content apps. Here's a summary of what's going on:

  1. Integration of AI-generated content (AIGC) services into its PC and mobile browsers

  2. Implementation of a "Shorten" button that can generate short summaries of webpages or articles

  3. Development of an AI-based solution to filter through all the content and summarise it

  • ๐Ÿ”— Link: Amazon's Monitron uses machine learning to help Amazon fulfilment centres reduce downtime by 70%. This is one fierce name.

Cool Creations

  • ๐Ÿ”— Link: can generate infinite texture variations for a given 3D mesh via a text prompt. In shortโ€”text-driven 3D texturing with AI.

  • ๐Ÿ”— Link: Google's Noise2Music allows for generating high-quality music audio from text prompts. In the not-too-distant future, you will be able to create any type of music you want via text prompts.

  • ๐Ÿ”— Link: AlesioRFM on Reddit is using Instruct GPT to show anti-clickbait summaries on YouTube videos. The number of thumbnails that get me to click is insurmountable so I'll be finding this useful.

  • ๐Ÿ”— Link: RunwayML introduced Gen-1, a new AI model that uses text or images to generate new videos out of existing ones. This can create video in any styleโ€”video-to-video is mindblowing ๐Ÿคฏ

Top Tweets

  • ๐Ÿ”— Link: Balaji dropped a thread on how AI directly threatens the income streams of doctors, lawyers, journalists, artists, professors and teachers. I feel open source does a great job of making AI accessible to everyone. Stability AI and Hugging Face are pushing this arena forward.

  • ๐Ÿ”— Link: Steve Nouri's tweet on the "AI Game of Thrones" shows how the competition for generative search is heating up. Who are you placing your bets on?

  • ๐Ÿ”— Link: Bing's new AI can summarise a 15-page PDF in seconds.

Startup Scene

  • ๐Ÿ”— Link: ElevenLabs raises $2M pre-seed led by Credo, a Czech VC for AI voice technology. They can produce artificial voices but also clone existing ones.

  • ๐Ÿ”— Link: Magic raises a $23M series A to build an AI-driven tool to help developers write code.

  • ๐Ÿ”— Link: Gloss raises $8M seed round to augment content creation from video. (Their promo video is awesome.)

  • ๐Ÿ”— Link: Kognitos raises $6.8M seed round to build an AI platform designed to automate business processes.

Weird DALLE of the Week

This one's a little scary, but I played way too much Minecraft as a kid not to share it.

Real-life Minecraft guy

Today I've got a question for you fine people. Twitter is my no.1 medium for sharing what I learn on deeper topics like understanding what GPT means in ChatGPT. Usually, these are reflected in the newsletter each Wednesday.

But I'm thinking of doubling down on these Sunday issues to make them the best they can possibly be with my own voice laced between the resources.

What do you prefer

Login or Subscribe to participate in polls.

I'm excited to hear your thoughtsโ€”I appreciate your feedback each week.

P.S. I'm looking for someone to help me set up and moderate the Discord channel. Reply to this email and I'll get in touch ๐Ÿš€

If you liked this piece, subscribe below: