Ben's Bites
Posts
2x $1.5BN Valuations

2x $1.5BN Valuations

Easily generate prompts, Excel formula generation, $101M seed round for Stability.ai, $125M seed round for Jasper.ai, 4k Stock video with music

Ben Tossell
October 18, 2022

Thanks for reading! Ben’s Bites is a roundup of what’s going on in AI. Not subscribed? Do that here.Link overload? Refer 2 friends with this link and get access to the filterable+searchable database.

Two AI companies valued at $1.5 Billion. Stability AI (makers of Stable Diffusion) and just announced Jasper - writing tool, raised $101M and $125m, respectively. These are astronomical numbers! We've hopefully got an interview with Jasper CEO coming this week.

I got access to Lex, the writing tool I talked about yesterday and used it to write this email. Although I don't use any tools to actually write the snippets. So I like it but not using its AI potential yet.

I played around with Play.ht which was behind the fake podcast stunt. So you can listen to today's email here. Shoutout to our fake narrator, Oliver.

Let's get to it.

Prompt: List the topics in today’s email:

📝 Easily generate prompts🔢 Excel formula generation💰 $101M seed round for Stability.ai💰 $125M seed round for Jasper.ai🎥 4k Stock video with music

🫦 BEN'S BITES

Image → prompt. We’ve all done it the other way around. Now you can generate a prompt by feeding it an image (Stable Diffusion). Shout-out to Bens Bites reader Clayton for surfacing this.

Comedians Tom Segura and Bert Kreischer have a podcast 2 Bears, 1 Cave and in their latest episode, they use AI to generate portraits of each of them. They’re pretty funny!

Another tool to help you with generating prompts! I saw this demo a few days ago and said it should be a product for others to use, and to his credit, Michael got it done. It’s live and this is what it looks like:

Runway, the video creation and edition (magic) tool has released: Erase and Replace. It allows you to swap out images in your video, seamlessly.

👋 Too many links?! I created a database for all links mentioned in these emails. Refer 2 friends using this link and I'll send over the link database.

Got an iPhone? You can now become a 3D artist. LumaLabs (mentioned in past emails) gives you the ability to export your captures as diffuse 3D assets!

For all you bookworms, this guy has dropped Chapter 1 of his book. Topics include; 7-part history of AI risk, effective altruism, the rise of rationalism, demon-haunted cults, and the birth of Bitcoin. New chapters are released daily!

This article has been making the rounds (I should’ve included it in yesterday’s email, whoops). Essentially the points made are that Artists say they work for years on their portfolios and now people can now make copycat images in seconds. But some AI companies argue that the new artworks are unique and can be copyrighted. What’s your take?

This thread discusses different use cases for using AI tools. It includes; overcoming writer’s block, outline generation, re-writing/enhancing content, ad copy, and more. Related thread from people who actually use these products and what they use them for.

Is this imagination? A reddit user created a character and asked AI to imagine stuff and wrote how AI could have its own imagination. Related phenomenon - The AI effect.

Another writing tool that I was linked to - Moonbeam

Imagic: Text-Based Real Image Editing with Diffusion Models. Basically, it can let you change the posture and composition of one or multiple objects inside an image while preserving its original characteristics.

PhotoRoom is able to generate marketing images for any product using stable diffusion. The example is: capture this car toy & prompt “a photo of this car driving a scenic mountain road”.

DiffLinker is now on Hugging Face. Given a set of disconnected fragments in 3D, it places missing atoms in between and designs a molecule incorporating all the initial fragments.

AI created a 4K stock video with music!

Accenture is hiring AI interns.

Sequoia Capital has created a Gen AI Market Map. It’s a living document because the space is moving so quickly.

A demo for Stable-Dreamfusion: A pytorch implementation of text-to-3D dreamfusion, powered by stable diffusion.

Jordan Singer also puts out impressive products, he’s releasing a magical design tool powered by AI for Figma.

Waverly launched to cut out the noise of the internet. Distraction-free content consumption.

The Palace of Fine Arts was altered with AI art.

The GitHub Copilot story is developing…Matthew Butterick is investigating what’s going on with what code GitHub is using and the legal complexities.

Excel Formulas - One of life’s many mysteries…But now, you can type what you’re actually looking to do and this tool will translate it into an Excel Formula for you!

A new AI model can accurately predict human response to novel drug compounds.

Vaccines to treat cancer, possible by 2030, say BioNTech founders

People are always asking how to build a successful company, so Nat put the question to AI. Here’s the advice it came up with.

CommonSim-1 allows you to render 3D images of any object (or use the library of uploaded assets) and create anything you like, in 3D. You could renovate your home or office space by seeing which items work and in which areas.

LAION-5B: An open large-scale dataset for training next-generation image-text models.

An AI dream journal - AI images based on this redditor's real dreams

💰 FUNDING ROUNDUP

Stability.ai raised $101M funding led by Coatue and Lightspeed.

Jasper.ai raised $125M funding led by global software investor Insight Partners and joined by other leading firms including Coatue, Bessemer Venture Partners, IVP, Foundation Capital, Founders Circle Capital, HubSpot Ventures and more.