17th issue: OpenAI has been busy this week 🏗️

🌯 Wrap for the week

A close look at all the main headlines in AI

There is a lot of news from OpenAI this week

Here are the top 3 news from them this week:

1. No more waitlist for GPT-4 API
   Get your code hats ready because GPT-4 is now available via its API! Not everyone gets to play with the image-understanding function yet, OpenAI's still testing the waters with a partner. With a bit of patience, developers will get to fine-tune this fancy function with their own data later this year. Also, a quick note to say "bye" to GPT-3 as it’s gonna be gone starting in 2024.

2. Code Interpreter has arrived for Plus users
   My wish to access Code Interpreter has finally been granted by OpenAI this week. This handy feature allows ChatGPT to run code and handle files you've uploaded, transforming everyone into a potential data analyst. Despite some grumbles, this upgrade offers an exciting glimpse into the future, promising less mundane tasks and more meaningful work for us humans.

3. Committing to search for alignment

   In light of the raising concern over AI, OpenAI is launching a new research division called "superalignment" to prepare for the rise of superintelligent AI. Such AI could pose significant threats if it starts to disagree with human values. Focusing on this type of AI, which is expected to exceed human intelligence, OpenAI will dedicate 20% of its compute power to address alignment issues. This new division, aiming to achieve the technical challenges of superintelligence alignment within four years, highlights the rising importance of AI safety.

Main news credit:
OpenAI makes GPT-4 generally available from TechCrunch
Code Interpreter comes to all ChatGPT Plus users — ‘anyone can be a data analyst now’ from VentureBeat
OpenAI commits to ‘superalignment’ research from TechMonitor

Microsoft’s new kid in the AI block

Cue the drums for CoDi, a new artificial intelligence prodigy from the Microsoft Azure Cognitive Service Research and UNC NLP teams. This generative model is not just a jack-of-all-trades; it's a master of them, capable of creating high-quality content across multiple domains—think text, images, video, and audio all singing in harmony.

Introduced in the paper "Any-to-Any Generation via Composable Diffusion", CoDi is more than just a pretty face in the crowd of AI models. It's part of Microsoft's Project i-Code, an initiative set to give multimodal AI a hefty push forward.

In the field of content creation, CoDi is eyeing the chance to become the Picasso of AI. From engaging social media posts to interactive multimedia presentations, it's poised to change the way content is created and consumed. As AI continues to evolve, CoDi showcases a future where human-AI interaction is as commonplace as your morning cup of joe.

Check the demo of CoDi 👇️ 

Main news credit:
CoDi: Revolutionizing Education - Microsoft’s Innovative AI Model Leads the Way from Mashable

Other top AI headlines for the week

Japan emphasizes students' comprehension of AI in new school guidelines from the Japan Times

Talking about embracing AI, Japan is going all-in on this. If you ask me, I think OpenAI would open its second international office in Japan.

Google Says It'll Scrape Everything You Post Online for AI from Gizmodo

Sneaky, sneaky Google. I understand you need content to train AI, but scraping everything is quite a stretch.

Harvard Will Teach Students Using an AI Instructor Next Semester from Futurism

I am not sure what the students might think about this approach, but overall speaking, this should help reduce the workload from the educator while providing a more tailored experience for students. Do you think this is a good idea?

🎬️ Guess that movie

Adding a bit of fun to your weekend reading, let’s guess the famous movie from the 10-emoji description created by ChatGPT:

🦍🌕👨‍🚀🚀🛰️👁️🤖⏳👶🌌

If it is a bit hard to figure out, you can find the answer with some explanation from ChatGPT at the end of this newsletter.

🧰 Toolbox

Check out this week’s tools and resources that give you more superpowers with prompt crafting!

1. 🔧 Playground AI
I haven’t used this before but there are quite a lot of features from this new AI graphic generator and editor. Also, they have recently raised $40M. So I guess they do have some money to spend to make it even better. Give it a try.

2. 📺️ LangChain: Chat with Your Data
Another great production from DeepLearning! Check it while it is still available for free now.

3. 📚️ The Rise of the AI Engineer
The article contains a good discussion regarding the role of AI engineer and how it compares to other existing roles in the market now. Take a look if you to check out this author’s perspective.

👋 Your next Twitter friend

Connect with great people in the industry to supercharge your growth

🖼️ Image of the week

While AI has been the focus of this newsletter, if you have been following the tech news, the no. 1 topic this week must be the crazy growth from Threads.

Well, let’s see what the combination of Mark and Thread would give us from Midjourney

🎬️ Guess that movie - answer

The answer is "2001: A Space Odyssey" 🥳 

Did you guess that right? Here is ChatGPT’s explanation:

🦍🌕👨‍🚀🚀🛰️👁️🤖⏳👶🌌

This summary captures the main elements of the story: prehistoric apes (🦍), the moon (🌕), astronauts (👨‍🚀), a rocket or spaceship (🚀), a satellite (🛰️), the HAL 9000 computer's eye (👁️), artificial intelligence (🤖), the passage of time or evolution (⏳), the Star Child (👶), and the universe or space (🌌).