Click HERE to watch video.
I did not write this. ChatGPT did.
Unless you've been living under a rock, you've been hearing a lot about ChatGPT and synthetic media lately. I decided to see if ChatGPT could write this blog for me, and if synthetic media could create my video. To do this I logged in and asked it to write a 700-word blog post about using ChatGPT to write blog posts.
Here's what it returned. Please remember I did not write this:
“To use ChatGPT to write a blog post, start by providing the model with a prompter topic you want the post to be about. You can then use the model's generated text as a starting point for your post and edit it as needed to make it more accurate, engaging, and interesting.
Using ChatGPT to write a blog post is a simple process that can save you a lot of time and effort. Here's a step-by-step explanation of how to use ChatGPT to write a blog post.
- Before you start writing, it's important to choose a topic for your blog post. This will be the prompt that you provide to ChatGPT to generate text.
- Access the ChatGPT model by using Open AI's API or other access point.
- Once you have access to the model, you can send the prompt and receive generated text in return. The model will use its training to generate coherent and grammatically correct text based on your prompt.
- The generated text is a great starting point for your post but will likely require editing to make it more accurate, engaging, and interesting. You can edit the text to add more information, correct inaccuracies, and make it more in line with your writing style.
- Check for plagiarism. As the model is trained on a large data set of text, it may generate text that is not unique.
- Publish. Once you've finished editing your post and fact checking the information, you can publish your blog post.
There are several reasons why someone might choose to use ChatGPT to write blog posts:
- Time saving. Writing a blog post can be a time-consuming task, especially if you are struggling to come up with ideas or find the right words. ChatGPT can generate coherent and grammatically correct text, providing a starting point that you can then edit as needed.
- Improved productivity. Using ChatGPT can also help improve productivity by allowing you to generate multiple versions of a post quickly.
- Consistency. ChatGPT can be used to write a blog post on a consistent basis and can help maintain a regular blogging schedule.
- Variety. With a wide range of topics, it can help generate a variety of posts, which can be helpful for bloggers that are looking diversify their content.
- Language proficiency. ChatGPT can be used to write blog posts in multiple languages. This can be helpful for bloggers or content creators that are not proficient in a particular language.”
I did not shoot the video ChatGPT wrote. RenderMedia AI did.
Okay, now me again.
Next, I uploaded the text to RenderMedia and had them create the video attached to this blog using synthetic media AI, otherwise known as deepfake technology. I also asked ChatGPT about the implications of deepfake technology. Here is what it wrote:
“Deep fake technology has been in the news due to its increasing use in creating realistic videos of people saying and doing things they never actually did.
One high profile example of this was the deep fake video of former President Barack Obama, in which his face and voice were manipulated to make it appear as though he were saying things he never said.
Another example of deep fake technology being used in a negative way, was a deep fake video of a well-known actress where her face was superimposed on a pornographic video in a non-consensual way.
It's worth noting that the use of ChatGPT and other AI-powered text generation tools can be controversial, particularly when it comes to issues such as plagiarism and the spreading of disinformation. As a result, some users may be hesitant to admit that they are using these tools. Additionally, as the use of ChatGPT is becoming more common, it's also becoming more difficult to detect if text was written by a human or by an AI model.
On the positive side, combining ChatGPT text with synthetic media video, can create highly engaging and interactive content. It can be used to create virtual presenters or hosts for new shows that can interact with guests in real time, making the show more engaging and interactive. Additionally, it can be used to create videos with realistic lip sync and speech, which can be used in the entertainment industry, such as movies and TV shows.
Combining ChatGPT text with synthetic media video has the potential to create highly engaging and interactive content but it also raises serious ethical and legal concerns, such as spreading disinformation, harassment, and impersonation. It's important to be aware of these concerns and to use the technology responsibly, ensuring that it's not used to harm or deceive individuals.”
Hey, it's Bruce again. Remember, I didn't write any of this and I didn't actually read it on camera, I simply uploaded the text. Imagine the implications…
Oh. My. Holy. What. The. Heck. This is terrifying and amazing and 1984 all at once. Great post man.
Bruce, This is so well done and really explains what is happening. the Good the bad and the ugly. It really is a game changer on so many levels. Thank you for sharing this.
Just saw your latest video and email on AI. Pretty remarkable. Indeed, the fact that a portion of your video was created by AI is mind boggling.
I saw the Obama video. Very scary and after watching it I took it for granted that NSA or the CIA or some spook agency has already created something that can determine authenticity of such recordings using metadata or other embedded fingerprints. All of which is consistent with my own theory that technology on this level actually can impair expedition as it constantly requires new failsafes and spawns active competition which often serves up improved products slowing down consumer selection.
A lawyer friend recently opened his own shop where he employs Watson as a “timekeeper” handling billable matters for clients. Three years ago I read reports that Watson was 70% as effective as a lawyer. I suspect that has only gotten better since. Predictably, coming soon will be the war over assignment of liability for what AI got wrong and was not detected/edited by the vendor, including the viability of the likewise predictable contracts that try to shield the vendor from that liability.
It gets especially tricky when the vendor is a professional, particularly a lawyer, whenever there is a collision between informed consent about work performed by a non lawyer, on one hand, and professional standards of care and public interest, on the other. I believe the former generally loses to the latter. Right outcome, I think, but it can be ameliorated by a well-defined review process, perhaps.
Anyway, fascinating stuff and your post dramatized how far along AI has come. Always enjoy your thoughtfulness.
It is scary to think that a machine can do exactly what a human can do in writing and speaking. It’s 1984. Not sure of the implications. This could allow a poor student to become a superstar without even trying.😫
Bruce!!
I have been chasing Deepfake for nearly three years!! Such potential!!
It started with Leslie Stahl’s interview with a block of videos on 60 minutes:
https://www.youtube.com/watch?v=D9tZnC4NGNg&t=38s
Then, during the Fall of 2020, I read The Future is Faster then You Think by Peter Diamandis and Steven Cotler. I could only read a couple pages at a sitting in order to keep my head from exploding.
Early part of 2021, an idea gelled. I would use an avatar on my website to help visitors through their property searches. When you map out the dialog flow, it is all fairly predictable. I started searching for AI developers and found several in India that actually answered my inquiries.
It is an iterative process, and has been going on for about 18 months. The voice cloning part was (back then) fairly expensive and required a subscription, so, in the meantime, I have recorded hundreds of audio clips in my own voice and we are at the point where the developers are able to craft a single response from my avatar made up of several clips.
The avatar resembles me, but is clearly not.
The purpose of this? I have a fair amount of luck with people knowing, liking, and trusting before meeting me. I saw this as an opportunity to amp that up a bit.
Exciting times–Thank you for posting this today!
Bob
Interesting times, but we will handle it. Cambridge Analytica’s work in 2016 was our warning shot. Distrust in media will finally reach its pinnacle and we may end up back in a simpler time. Academia, which is really being challenged as students are typically ahead of professors and administrators. A realignment is on the horizon. Time to get back into the real world of showing up in the room where it happens.
Namaste~
Bruce,
For the record, I tend to gravitate early to new ideas. Always been that person who fast forwards past the first 2 stages of Schopenhauer’s 3 stages of truth. Things tend to be self-evident to me right away.
–
But this ChatGPT shit actually scares the crap out of me. At the onset, I cannot find one semblance of good that can come from it. Helping people who can’t write or can’t market doesn’t create a better world for anyone.
But more importantly, what frightens me is the may ways this can be weaponized. In a world where I could receive an email from you that you never sent me – a crime that happens to all of us at some point – what would stop some sinister hacker from creating a compromising, bogus post and video in someones name and publishing it out in the world without that person knowing?
How would any of us even defend ourselves against it?
I’ve been tempted to test this tech out given my curiosity about all things (I bought Bitcoin for the price of Apple Dongle despite not really understanding what the heck it was), but my Spidey senses are screaming at me to run far from this stuff.
I dunno, what am I missing here?
I honestly can’t think of one good deed this technology could be assigned to which means if there is, I am on aware of it and can use some education on it. For edification, I don’t believe helping someone who can’t write or market is the better world, good deed I’m hoping for.
The implications?????? Frightening! How do we know what is real?
Definitely & absolutely scary…..I am going on record to say, this is 1984 and the Terminator is becoming more and more a reality….the next 10-20 years should be interesting for the human race…
Oh and I did write this, just check out the grammar, LOL
Stop the world – I want to get off…