Thanks for going deeper with us!

Happy New Year! It’s finally time to circle back to everything you put on the back burner in December.

In today’s digest:

• Rounding up some of the biggest AI launches of 2023

• Google just launched a new text-to-video AI

• Scientists and AI have teamed up for the sake of our health

And don’t forget to tell us what you thought about today’s newsletter at the end!

— The Product Hunt team

P.S. Someone awesome forward you this newsletter? Sign up here.

WHAT A YEAR

2023’s biggest AI launches

It’s officially the new year. 2023 has come and gone, but what a year it was for AI.

It felt like every day, some new groundbreaking model dropped to revolutionize the world. For some, it felt hard to keep up with, but lucky for you, you read Deeper Learning 😉.

Now that we are in a new year, it seems right that we look back on some of the biggest AI launches of 2023.

GPT-4: Open AI dropped GPT-4 back in March of 2023 as a successor to the GPT-3.5 model. The company claimed it demonstrated human-level performance across several benchmarks, and while it hasn’t lived up to all of the internet’s QA testing, it has become a staple in many people’s productivity stacks.

Bard: Seeing the writing on the wall, Google jumped head-first into the AI water to catch up and launched Bard — its answer to ChatGPT. Google launched Bard initially as an experiment, but it became progressively more feature-rich throughout the year.

Grok AI: Google wasn’t the only one feeling FOMO. Elon Musk launched Grok AI for X (formally Twitter). With a name that was inspired by the Hitchhikers Guide to the Galaxy, Grok is a chatbot with a more sarcastic and humourous tone.

Midjourney V6: Arguably, V6 is the most capable Midjourney update yet. It launched last month and fixed several issues that plague AI images, such as hand generation, text, lights, and shadows. The images Midjourney now produces are so much more realistic that the team had to ask users not to use them in bad faith.

GPTs by OpenAI: For much of last year, indie makers built their own ChatGPT wrappers for different purposes. Open AI noticed and decided to capitalize on it by allowing anyone to fine-tune an instance of ChatGPT for their own purpose, and even sell these new personalized ChatGPTs in OpenAI’s own version of an app store.

What will 2024 hold? AGI? It’s anyone’s bet!

But you can see more of the top AI launches from 2023 here.

DALL-E KILLER?

Google just dropped a new video-based AI model

Google is coming out swinging in the new year with its latest model: VideoPoet. VideoPoet is the latest addition to Google’s promised “multimodal AI,” meaning it can generate text, audio, images, and even video. The last part is where VideoPoet fits in.

VideoPoet is an experimental large language model (LLM) that’s capable of a number of different video-related generation tasks, including text-to-video, image-to-video, and even video-to-audio. 

Take text-to-video as an example. You can input something as ridiculous as a “raccoon sipping wine in Times Square,” and within a few seconds, you will have the generated video ready to be shared in whatever style you requested. Just imagine the Mona Lisa lip-syncing an ABBA song.

Already have an image you want animated? Drop it in, and VideoPoet will generate a short clip based on the image. Once the video is generated, you can edit it further and stylize it to how you want just by telling the AI your instructions. 

Check it out

WOAH!

Science 🤝 AI

To you or me, AI might mean chatbots, image generation, and videos from text — pretty mind-blowing stuff. But one of the most important areas where AI is changing the world is in the health sphere, like in the fight against antibiotic resistance.

A team of scientists discovered a new compound that could potentially help them craft the first new antibiotics in over 60 years. The compound is suspected to be able to neutralize a drug-resistant bacterium that kills thousands of people every year.

AI was apparently a game-changer in this discovery. “The insight here was that we could see what was being learned by the models to make their predictions that certain molecules would make for good antibiotics,“ said James Collins, professor of Medical Engineering and Science at MIT.

UNTIL NEXT TIME

Written by Aaron O’Leary with Sarah Wright.

Here via forward? Subscribe here.