I've set *most* of my videos to Public again

Hello everyone.

Many of you have been asking for me to put some of my best videos back online. Today, I’m pleased to announce that most of them are back online! This includes most of the original AutoMuse videos, as well as many of my finetuning projects, as well as a few other videos such as my “Answer Complex Questions from Multiple Documents” video.


Why did I take them down in the first place?

When you look at the present backlash against AI art, you have your answer. I realized, before the rest of the internet, that the work I was doing had destructive and harmful potential. Most of my closest friends are writers and the moral question was simple:

What would my friends want? What would hurt them most and least?

The calculus was simple. If I built a tool that undercut their work, I very well could lose my friends. Because of this, there are several projects that will never see the light of day. That being said, it’s entirely possible others will figure it out.

Okay, so why put them back up now? What changed?

Quite a bit has happened since I took down my videos. Most importantly, I’ve been talking to many people. I’ve spoken with industry veterans, folks who saw the Dot Com Revolution (and subsequent crash), I’ve spoken with professional and semi-professional authors, and I’ve also spoken to industry insiders.

Second thing; ChatGPT happened. ChatGPT changed the conversation around AI-based tools. While ChatGPT has its flaws, it also has some strengths and some cool potential. ChatGPT is superior than some of my tools, but there are still things I can do that ChatGPT cannot, and those are the projects that will stay suppressed for now.

Third thing; my focus has shifted. As I’ve come to understand the market (supply and demand for AI tools and services) as well as the zeitgeist (excitement and backlash around AI) I have come to realize that 99% of all ventures in the AI space will fail. Everyone is building shiny new tools that will be irrelevant within months. We have never seen anything like AI and what’s coming. I recently made a video about some of my predictions for 2023, but don’t take my word for it.

Microsoft, providing OpenAI capital and computing power for its software, is a beneficiary. Asked about ChatGPT and whether Microsoft viewed such technology as experimental or strategic, its President Brad Smith told Reuters that AI has progressed faster than many predicted.

"We’re going to see advances in 2023 that people two years ago would have expected in 2033. It’s going to be extremely important not just for Microsoft’s future, but for everyone’s future," he said in an interview this week.

I will also be setting some of the code back to public, but it will take a while to go through and clean it up. Check out my code here:


David, this was a good move.
I was a bit disappointed when you took them out. As a researcher (as I understand you to be), I think the value is in experimenting and putting stuff out. What people do with it is beyond your (and mine/ours) control. No one knows where things will be headed, or who will be harmed or not.

Let’s focus on the positive things and keep moving forward. Like you said, there’s a lot of hype and most will be irrelevant and fail.

1 Like

Yeah man, I know a lot of people were bummed. I know that it was a setback for many people but I’m still glad I did it. It’s better to be cautious and know for certain you’re doing the right thing.

1 Like


Thanks for sharing. I noticed this one is not available (or I can’t find it). Could you share some insights of what you achieved. I am mostly interested in knowledge graphs. Argument mining is hard, ontologies are key.

1 Like

Yeah sorry it is getting some updates. It will be back online within a few hours. It needed editing and HD processing. Stand by…

1 Like

It’s fixed and back up.


Thank you Dave! I love watching your videos. There’s a lot people can learn from them.

1 Like

Saw the time and date (on your screen) and only wished to send you coffee :slightly_smiling_face: Can’t wait for part 2.

I am currently working on similar stuff. I am referencing below a couple of medium articles (there is code, a colab notebook) I was required to read a few days ago.

Do not know how you are going to make the relationships, but here are two resources.

Extra: a nice reading on Network Analysis Visualization. This guy did a piece on Game of Thrones and had quite a few good rules for deriving relationships.

1 Like

Thanks man, this is super helpful. I will go over these during tomorrow’s video as part of the extraction process. Visualization will be part 3, I think

1 Like

Part 2 is up


This project is done, not quite successful. However, it gave me a lot of new directions to go in. I will be exploring CI/CD, Agile, and a few other aspects of high velocity development with ChatGPT next. This includes identifying problems with code, developing user stories, and adding new software capabilities entirely with ChatGPT. I’m just steering the ship.


Thank you for the channel, I have been watching a lot of your videos and just finished this knowledge graph series. I have been considering doing something similar in my country but with skills development legislation to better understand industries and student alignment. Needed a way to work with all the pdfs and found these videos helpful. Still have a lot to do though. Also love going down the AI rabbithole.