Good morning.

We have a feeling there’s BIG AI news coming this week.

But before that, even BIGGER news. We launched our Beginner’s Guide to Midjourney.

Peep the links section to check it out.

🤖 Top Stories

😬 Most Americans have these hot takes on AI

Americans are still verrrrryyyy skeptical of AI.

They also don’t believe what it says is actually true.

538 analyzed quite a few polls about AI. Most come to this conclusion.

Key takeaways:

1. A majority of Americans thing that generative AI isn’t a fad

2. 80 percent remain undecided on whether output from generative AI is trustworthy

3. 75 percent think health care providers would move too quickly adopting AI

4. 78 percent of Americans think AIs writing news articles is a bad thing

5. Two-thirds of adults worry about misinformation on AI search engines.

Wooof. That is rough.

What’s the silver lining?

What do they think AI is good for?

Recipes, roadside assistance, and coal mining.

The “big three” as they say.

Why does this matter?

Clearly, Americans either don’t know much about generative AI or are quite skeptical.

As they say, we’re still early.

⚡️ Qualcomm proves generative AI belongs on your smartphone

The Intel of mobile phone chips set an AI record.

They showed off the fastest-ever deployment of an AI image on a smartphone. With the help of Stable Diffusion.

The demo is impressive. It shows a Qualcomm-powered phone generating the following prompt in 15 seconds.

super cute fluffy cat warrior in armor, photorealistic, 4K, ultra detailed, vray rendering, unreal engine.

Not only is it a speed record. Image generation like this almost always takes place in the cloud. It takes a lot of computing power to render these images. Whenever someone is using Midjourney, for example, the request isn’t run on their computer. It’s in a GPU in a data center somewhere.

Qualcomm did mention its engineers optimized the model for the test, but that doesn't change how impressive this demo is.

Why does this matter?

AI is more accessible if it can be run on local hardware.

Generative AI in every phone on the planet sounds not too far off.

👀 OpenAI outlines how to avoid robot takeover

Clearly, all these AI advances have people spooked.

So OpenAI did us all a solid and published their guidelines on AGI for when the robots do eventually rise up.

First off.

What's AGI?

AGI refers to the development of a machine that can perform any intellectual task a human can.

Think of it like a supercharged human that can learn faster and better, but it’s a robot.

Once this happens, some predict an AI apocalypse, while others predict a utopian future augmented by AI.

OpenAI wants to make clear that it shouldn’t be to blame if the bad stuff happens.

They outlined three key questions when it comes to AGI.

1. How do we govern these AI systems?

2. How do we ensure their benefits are widely shared?

3. How do we mitigate risks associated with their development and deployment?

In English: Who is in charge, who benefits, and how do we ensure nothing bad happens.

What’s most interesting is the second point. It’s pretty clear that unchecked AI can benefit developed nations first. How do you ensure equal access, when there isn’t equal access now?

Why does this matter?

OpenAI is the biggest AI player in town right now.

Long-term, they want to make clear that the future of humanity is for…humans.

And that means careful consideration over what AI actually controls.

🤖 Special Feature

The Ultimate Beginner’s Midjourney Guide

If you’re learning Midjourney, we created the ultimate guide to help you supercharge your AI art. Grab it at the link below and let us know what you think.

😎 Cool links

1. Logos through Controlnet.

2. Hold your horses on MetaGPT.

3. AI rap battle.

4. Personalized meme generator.

5. Images upscaled, using AI.