This AI tool grew to over 1 million users in only 5 days

Hello!

Welcome to the first-ever issue of the Practical AI newsletter. A newsletter free from geek-speak, where complex terms are simplified and broken down into digestible bits even my mom would understand. 

On the 30th of November 2022, OpenAI released its ChatGPT, and I had the biggest epiphany of my professional life. It was as if I got sucker punched in the face by Mjölnir (the hammer of Thor - the Thunder God).

AI will affect everything. Business, work, and life as we know it! The aim of this newsletter is to introduce you to, and educate you on the fantastic (and scary) capabilities of Artificial Intelligence.

Newsletters are a brand new thing for me, so this newsletter will probably evolve as we go. However, this is what I want to share with you weekly: 

• AI news

• AI tools

• AI applications

• Interviews with AI founders

• Show practical examples of AI in use

• Cool and crazy stuff

Any feedback is greatly appreciated! Just hit reply, or tweet me @thomassorheim 

Who am I? My name is Thomas Sorheim. I am a Norwegian currently overwintering in sunny Portugal. Although I am not working in the AI industry, I have a background in tech, e-commerce, and digital marketing. 

With that out of the way, let's jump into the heart of this week's newsletter.

ChatGPT from OpenAI grows to 1 million users in 5 days

It is fitting to start the first issue of the Predictable AI newsletter by talking about ChatGTP. It was the technology that gave me the epiphany of my life and made me understand that AI is going to be a massive part of our lives in a matter of a few years.

On top of that, as a digital marketeer, I understand who, or rather what, will disrupt the all mighty G - Google that is….

It took Spotify 150 days to get to a million users.It took Instagram 75 days to get to a million users.It took ONLY 5 days for ChatGPT to reach a million users. 

Source: https://read.chartr.co/newsletters/2022/12/9/ai-goes-mainstream 

This is HUGE! 

The tool crumbled when the general public caught on, which was basically this week. 

I am unable to log in and use ChatGPT most days. 

On Monday, I was met with this error message!

Although ChatGPT is in its infancy, AND the algorithm is not connected to the Internet, and its data has not been updated since 2021, it is still very impressive! 

It gives us a glimpse into the future. 

The director of OpenAI, Sam Altman, warns us that the tool is limited but good enough at some things to create a misleading impression of greatness. 

As much as this is true today, it is just a matter of time before the AI and the technology have evolved so much it is ready to be taken to the next level. 

For the fun of it, I asked ChatGPT what GPT-3 is (the AI brain of ChatGPT). It is like you ask someone, “Who are you”, or in this case, “What are you”.

Right now, we should regard ChatGPT as a novel tool with a limited set of use cases (especially since we cannot really trust the data 100%). 

However, here are a few things I have used it for so far: 

• I asked for a good night story for my daughter (read here).

• I asked it to write a poem about my daughter (see below).

• I had a discussion with it about religion (as an atheist). Unfortunately, I deleted it by mistake…

• I asked for a recipe for Indian Butter Chicken and how to make it (read here).

• I asked for name suggestions for a newsletter about AI (I chose Practical AI).

• I asked for name suggestions for a blog about AI. ETA is next week, so you have to wait until then for the grand reveal.

• I ask it all the time for dad jokes, and it always delivers (as you will see at the end of the newsletter).

• I asked it for frequently asked questions people have about AI. I will use these to write answers and articles on the AI blog TBA. 

• I asked why fast food is not good for me, in the voice of Shakespeare (see below).

The poem for my daughter Martha

Fast food warning from Shakespeare

If you want to see some of the crazy stuff AI tools can create, I created a separate website to display outputs. I will be publishing AI fun weekly, if not daily. Check it out MadeWithAI.org

You can sign up for ChatGPT here for free: https://chat.openai.com (it is still struggling, but give it a spin)

Applications of ChatGPT 

I don’t think AI technologies can replace copywriters (at least not yet). But imagine the time saved by asking an AI writing tool, “what topics should an article on Sharks cover?” and getting back an outline with the topics, suggested headlines, and in what order it should be in. 

The AI tool has already saved you a lot of time. Agree?

Then, once you have written the article, you can ask the AI, “did I miss anything?” and it will tell you. 

Then, you can ask another AI to tell you if you should add certain keywords in your text to make the article more in line with the articles currently ranking on the first page of Google (this is important for SEO and rankings).

Before you publish, you ask the AI to create images to complement the article. 

How many hours has the AI already saved you by following this process? 

I would guess at least two hours, maybe more! 

Is this real? Is this in the future to come? 

The above scenario is not from the future. 

It is already here. 

Tools like Jasper AI, Copy AI, Surfer SEO, Dall-E, Luna, and several others are part of the current AI toolset that can make content creation faster than only a few years ago. 

Implications? 

As someone who runs a few content-driven businesses (websites), I rely heavily on external writers. I cannot check if the articles I get from my writers are written by them or by an AI tool. 

I am actually at the mercy of their honesty.

And this can be a problem…

But the biggest problem is not being taken advantage of or being cheated. That just sucks!

The real problem is that if they use an AI tool to write the article, Google will find out, and I will be penalized and lose rankings in Google. 

Lost rankings translate to lost traffic, which translates into lost revenue. And lost revenue translates to all kinds of bad shit. 

AI creates efficiencies. But it will also bring a lot of challenges and change. Not all of these will be welcome - at least not in the short term (check out the Adobe article linked below). 

A fun little tidbit about creating content with AI

My friend Amando wrote a book about How To Write A Book With AI. 

It was written entirely with AI… 

AI image of the week: “Steve Jobs by Rembrandt.”

This image was generated by the tool AISEO Art, which you can buy for $59 for a lifetime deal over at Appsumo right now. 

You can find more fun images I created over at MadeWithAI.org 

Quick bites

Is the academia prepared for AI? Probably not. This is an interesting article from The Atlantic discussing AI and how technology is changing academia. 

Adobe is embracing AI and claims it can amplify human creativity. It is causing waves, as many digital artists see AI as a threat to their livelihoods. 

This copyright lawsuit could shape the future of generative AI. An interesting article on the legal implications of AI that is trained using existing art, text, and programming code.

AI definition of the week: Deep learning

Deep learning is a type of artificial intelligence that uses multi-layered neural networks to learn from data and make predictions. It is a powerful tool for complex tasks such as image and speech recognition, natural language processing, and decision-making. Deep learning has become an important part of AI research and has been applied to many different applications. 

(This was written by ChatGPT - I asked for a simplified definition)

That’s it for the first edition of the Practical AI newsletter. 

Any feedback is greatly appreciated! Just hit reply, or tweet me @thomassorheim 

PS! Why did the AI model go to therapy?

.

.

.

Because it was having trouble learning to cope with its own biases. 

This is not the end, it is where the fun begins.