/w=32,quality=90,fit=scale-down){kind=small}
Raphael ThysGenerated from Dall-E by OpenAI
In today’s digital age, more and more content is being created by artificial intelligence (AI) rather than being written by human authors. This has led to a debate about the value of AI-generated content compared to hand-written blogs. While AI-generated content has the advantage of being able to produce large amounts of content quickly and consistently, hand-written blogs offer a level of personalization and human touch that cannot be replicated by machines. Furthermore, the quality of AI-generated content can vary depending on the quality of the algorithm and the data it is trained on, whereas hand-written blogs are the product of individual creativity and expertise. Ultimately, the choice between AI-generated content and hand-written blogs depends on the specific needs and goals of the content creator.
The entire paragraph above was generated by AI (via ChatGPT), and my only contribution was the input prompt.
AI is invading creative spaces: what does this mean for content creators?
As recently as September, the NY Times published an article about an AI program winning an art contest. I found this interesting as it prompts the question: what is the purpose of art? If art is meant to be beautiful and thought-provoking, does it matter if it’s AI-generated or human generated? Or is art a demonstration of human competency, and its value is in our appreciation of the artist’s ability?
While I am not interested in starting a philosophical argument about AI-generated art, I am interested in seeing if AI could generate blog content worth reading. I used ChatGPT, and I was shocked by the quality of the result.
In this blog post, I am going to link you to an article I wrote that took a substantial amount of my time. Then I’m going to post a similar one generated by ChatGPT. If you read both of them, you can get a fairly good grasp as to what ChatGPT can do well, and where it won’t do well.
· What is ChatGPT? · Elliot versus ChatCPT: the ultimate showdown · What implications does this have on blog content creators? · How can you use ChatGPT to be a better writer?
If you like this post, please 👏 , subscribe, or follow me on medium or LinkedIn to let me know I’m on the right track!
What is ChatGPT?
Earlier this week, OpenAI released a free-to-use public version of ChatGPT. This AI program allows you to ask a question or make a request, and it will use AI to respond as best it can. It’s not a search engine, but a much more complex algorithm that tries to understand your question and create an accurate response.
There are many applications for ChatGPT, but starting with a simple example: a replacement for Google Search.
Imagine this in the hands of Buzzfeed content producers. All they need to do is ask ChatGPT for a “top 10” list, slap in some images, and they will have a complete post. I predict that ChatGPT (or equivalent) will play a heavy role in this type of internet content.
Let’s try something sillier:
It probably doesn’t take advanced AI to come up with this answer, but it’s a good demonstration of the flexibility of ChatGPT.
As you’ve seen from the intro and the examples above, ChatGPT is more than capable of creating direct responses to questions, especially when there is available data for it to learn from. But can it generate a blog post with unique insights?
Elliot versus ChatCPT: the ultimate showdown
I was recently inspired to create a blog post that explains the key points of self-improvement books using pop culture as a mnemonic device. I started by explaining Stephen Covey’s 7 Habits of Highly Effective People using Game of Thrones’s Tyrion Lannister.
To research this post, I watched the first seven seasons of Game of Thrones and reread Stephen Covey’s book. I spent a couple of dozen hours trying out different potential metaphors until I had settled on a set of examples I liked. I had someone help me polish the grammar, and I slapped on some graphics. I felt good about it.
The post got some traction, and I am inspired to write a second one. This time around, I want to try pairing up Thinking in Systems by Donella H Meadows and Star Wars by George Lucas.
Before writing the article, I ran this pairing through ChatGPT, and the results are awesome. Below is the blog post that ChatGPT generated. As a reminder to the reader:
Everything in the section below is generated by AI. I did not edit it in any way.
Thinking in Systems and Star Wars: A Comparison (by ChatGPT)
The book Thinking in Systems, written by Donella Meadows, provides a framework for understanding how complex systems work and how they can be changed. The popular film series Star Wars, on the other hand, tells an epic story of good vs. evil set in a galaxy far, far away. At first glance, these two might not seem to have much in common. However, upon closer examination, it becomes clear that the principles of thinking in systems can be seen throughout the Star Wars saga.
One of the key ideas in Thinking in Systems is the concept of feedback loops, which are the relationships between different elements of a system that can either reinforce or counteract each other. In Star Wars, we see numerous examples of feedback loops at work. For example, the Rebel Alliance’s decision to destroy the Death Star creates a feedback loop in which the Empire’s fear and anger leads to more aggressive tactics, which in turn fuels the rebellion and causes more people to join their cause.
Another key concept from Thinking in Systems is the idea of leverage points, which are the places within a system where a small change can have a big impact. In Star Wars, we see this concept at play when Luke Skywalker destroys the Death Star, despite being just one small player in the larger conflict. His actions create a chain reaction that ultimately leads to the downfall of the Empire and the restoration of peace to the galaxy.
Thinking in Systems also emphasizes the importance of understanding the interconnectedness of different elements within a system. In Star Wars, we see this principle at work in the Force, which connects all living beings in the galaxy and allows them to influence each other. The characters in the story must learn to understand the Force and how it works in order to harness its power and achieve their goals.
In conclusion, while Thinking in Systems and Star Wars may seem like vastly different subjects at first glance, they both explore similar concepts and ideas. Both emphasize the importance of understanding complex systems and the role of feedback loops, leverage points, and interconnectedness in those systems. By applying the principles of thinking in systems, we can better understand the world around us and make positive changes to create a better future.
I am not going to lie; as a technologist, this is impressive. In just a few seconds, ChatGPT was able to:
- identify the key points from Thinking in Systems,
- identify examples from Star Wars that exemplify those points, and
- write an intro, conclusion, and content covering (1) and (2).
Yes, the metaphors aren’t perfect, and the format feels formulaic, but what it created was unique and amusing. It is amazing that today’s software can perform a task like this in seconds. As an engineer myself, what the OpenAI team has done is truly impressive.
Am I saying that ChatGPT is now the replacement for all writers everywhere? Absolutely not, but it does have strong implications for the future of blogging.
What implications does this have on blog content creators?
Generated from Dall-E by OpenAI
Much like the blacksmiths of old, I believe content producers need to adapt their approach with the introduction of this new technology. I’m a bit cynical, but I think many blog posts are just lightweight, regurgitated content. They take simple insights and repackage them as new content. ChatGPT will overwhelmingly replace these content producers, as their work is easily automated with AI.
I believe this leaves us with two types of human-written blog posts that will remain in the future:
- Blog posts that produce new information
- Blog posts that are written from a valuable perspective
AI systems are trained off of models and then produce insights based on those existing models. They can be presented with an idea or a question and formulate a response based on millions of data points. However, that doesn’t mean they produce genuinely new information. Here’s an example:
What is new information? If Twitter goes down tomorrow, you could ask a human or AI to speculate the most likely cause, but it wouldn’t be capable of telling you what actually happened. That is new information. Once that information has been written down and posted via a Twitter blog, it can be incorporated into human and AI responses to other questions.
If your blog post generates new information, you will likely be safe in a ChatGPT world.
Why is some content valuable depending on the author? If you wanted to read about the rise and fall of FTX, then a post written by Sam Bankman-Fried (the infamous CEO of FTX) would be interesting, even if he is an untrustworthy narrator. A post written by ChatGBT would likely be more accurate, but that doesn’t mean Sam’s perspective would be of no value.
If you are writing content about your personal experience, and your audience believes your perspective is uniquely valuable, then ChatGPT is not a threat.
How can you use ChatGPT to be a better writer?
My recommendation for blog post writers: use ChatGPT to validate if your blog post produces new information or a unique perspective.
I used ChatGPT to generate a template for what an AI-driven blog post would look like for the content I wanted to write about. I consider it the “minimum bar” that I need to beat. I will need to focus on (1) generating new information and (2) providing my unique perspective if I want my next post to be more catchy than ChatGPT’s.
Next time you go to write something, feed your thesis through ChatGPT and see what it produces. If your insights are easily reproduced by AI, then you likely need to think more about the topic, or your topic isn’t that insightful.
In the case of this blog post, my perspective as a human writer is the part that creates value for you (the reader).
Wrap-Up
Whether or not AI-generated blogs diminish the contributions of human blog writers is a matter of debate. Some people argue that AI-generated content is a valuable tool that can help human writers produce more content in less time, allowing them to focus on more complex and creative tasks. Others argue that the widespread use of AI-generated content threatens to replace human writers and reduce the value of their work. Ultimately, the impact of AI-generated content on human writers will depend on how it is used and how the industry adapts to its presence.
And of course, I used AI to generate the conclusion. But honestly, it’s a pretty solid wrap-up.