Good morning human brains, welcome back to your daily munch of AI news.

Here’s what’s on the menu today:

• AI recap from OpenAI to Dolly 2.0 🤖 

• Your phone might be haunted by Snapchat’s chatbot 👻 

• Photo editing just got a whole lot easier 📸 

• OpenAI’s CEO says the age of giant AI models is already over. 📉 

P.S. Bot Eat Brain just passed 4,000 readers! 🤖 😋 🧠 

MAIN COURSE

AI recap from OpenAI to Dolly 2.0 and beyond.

Here’s a quick recap on the saga that’s taken us from completely closed-source models like GPT-4 to the latest open-source AI models that anyone (with a beefy enough computer) can run for free.

First, Facebook (aka Meta) announced LLaMA, an alternative to OpenAI’s GPT-4 that they wanted to restrict only to certain researchers they approved.

Barely a week later LLaMA got leaked by 4Chan users, meaning anyone could download LLaMA and use it themselves, with the small asterisk* that they might get sued by Facebook if they used it to build a business.

Then, some researchers from Stanford showed that any large language model (LLM) like GPT-4, which cost many millions to train, could be replicated for just a few hundred dollars using untrained models like LLaMA and having GPT-4 itself do all the hard work of training it. That project was called Alpaca.

Last week, we got Dolly 2.0, perhaps the world's first truly open LLM. Dolly is special because unlike LLaMA and other not-quite-open models, the dataset, dataset licensing, training code, and model weights are all open-source and suitable for commercial use.

On Monday we got an even more ambitious attempt to build an open-source dataset for training LLMs: RedPajama-Data, which has over 1.2 Trillion tokens worth of training data anyone can use.

Big data, big whoop right?

Well my precious human brains, that’s not all.

As of yesterday we now have MPT-1b-RedPajama-200b-dolly — a ”1.3 billion parameter decoder-only transformer pre-trained on the RedPajama dataset and subsequently fine-tuned on the Databricks Dolly.”

Phew, that’s a lot. Caught your breath?

Here’s what’s next: We now live in a world where anyone who wants a powerful AI model can quickly and cheaply create one.

This means big companies and governments will need to tread very carefully as they develop the next generation of even more powerful AI.

If they create something dangerous it will quickly spawn thousands of almost-as-powerful replicas.

BUZZWORD OF THE DAY

FROM OUR PARTNERS

AI-powered competitive intelligence

Scout generates bite-sized summaries of your competitor's latest product updates, news, and content – so you're always on top of what's new.

SIDE SALAD

Snapchat AI is free now

Last month we wrote about My AI, a GPT-powered bot released as a paid feature inside of Snapchat. At the time, My AI was only available to the 2.5 Million or so paying Snapchat+ users.

This week Snapchat made My AI free for everyone.

Currently, the bot can only respond to chats or snaps with a text reply, but the Snapchat announcement promises that it will soon be able to use image generation to generate snap replies as well.

A LITTLE SOMETHING EXTRA

Inpaint Anything: Segment Anything Meets Image Inpainting

Earlier this month we wrote about Meta AI’s Segment Anything Model, aka SAM, an AI model which makes it wonderfully easy to identify and isolate every object in an image.

Now, a research group based in China has combined SAM’s powers with Stable Diffusion to create a tool that can identify any object in an image and change it to whatever you want.

Check out this example of a Dog sitting on a bench:

Want to remove the dog? Done.

Replace the dog with a Teddy Bear? Done.

Replace the bench with a swing? Done!

Here’s another example. With this tool, one click, and a short text prompt you can replace that donut in your hand with a camera lens, or practically any other object your heart desires.

Cutting your ex out of old photos just got a whole lot easier.

See more examples or read the full research paper here.

MEMES FOR DESSERT

Here’s a workflow shared on Reddit that uses Stable Diffusion and MiniGPT-4 to generate funny wikiHow articles like this one:

YOUR DAILY MUNCH

Think Pieces

AI is already taking video game illustrators’ jobs in China.

OpenAI’s CEO says the age of giant AI models is already over.

MAD, China, and the Semiconductor Showdown.

Large, creative AI models will transform lives and labour markets (The Economist).

Startup News

Stable Diffusion is hosting an AI animation challenge.

Google combines rival AI units DeepMind and Google Brain into a single super-entity for AI research.

Research

Align your Latents: High-Resolution video synthesis research from NVIDIA generates some of the highest-quality text-to-video results we’ve seen yet.

Farm3D: turn 2D pictures of farm animals into 3D animation.

Tools

AI Playground by Vercel Labs: simultaneously prompt two AI models and compare the results.

Replicate Codex: Search, filter, and sort AI models. Find the right one for your AI project.

TextGPT: Text with ChatGPT right from your messages app

texttomap: Generate interactive maps from a text description.

ReRoom AI: Upload a picture of your room to explore various design styles.

TWEET OF THE DAY

This Twitter user is combining large language models and other AI tools to make commercially available robot arms autonomous for much cheaper than the custom state-of-the-art robot systems we’ve written about recently (which were already much cheaper than previous systems).

AI ART-SHOW

Until next time 🤖😋🧠