(Disclaimer: Profanity.)
Or software engineers for that matter.
One of the many rumors (or hypes) that built these LLMs up when they were launched into the wild was that they could replace entire content writing teams or any other type of knowledge workers. Obviously, over time, we realized that it was a load of bullshit.
But why?
In the past two years, I have discovered many things that are unique and exclusive to human creators which are impossible for machines to replicate which are discussed below.
The objective of this chapter is to motivate writers not to take any shortcuts while leveraging LLMs to elevate their content writing workflow by educating them on what makes them unique.
1. Breathe life into creations
A crayon art of a scenic view of a 9-year-old is more valuable than a pixel-perfect image generated with the flagship model of Midjourney. Yeah, the latter will have more details and will be, theoretically, more aesthetic to look at, but in terms of value, AI art is worthless.
The reason is quite simple — creators breathe life into their creations.
There is a reason we are attracted to human-created pieces of art, whether it is a LinkedIn post or an earthen pot. These art pieces have warmth and have the ability to tell a story that none of the machine-built things will ever have.
It is hard to describe this phenomenon in scientific terms but the difference is clear. Anyone who has compared AI-generated text and human-written content on the same topic knows this.
If I have to explain it tangibly, I will use terms like engagement level, persuasiveness, and value over time. AI has been around for a couple of years and most of the content created by it will never be consumed more than once.
Human-written content, on the other hand, is often revisited and admired. In fact, the value of human creations increases as time passes.
2. Multi-layered thinking
While writing content authors have to consider multiple factors. These factors or aspects are often interlinked and complex. Some of them include the message, CTA, and format.
Within the message, you have even more sub-factors like tone, length, and style. Similarly, the CTA can be of any kind — read a blog, download a PDF, or fill out a form. A good writer focuses on the right aspect when ideating, writing, and publishing an article.
LLMs can’t do that.
You can create a rule engine. A rule engine consists of a few algorithms connected to one another in a particular sequence. The first algorithm, in the context of content writing, can be “generating ideas”, the second one can be “creating the outline”, and so on.
Basically, you can create a sequence of detailed instructions that take your idea and convert it into a publish-ready draft. However, it will never work the way you want it to. Writers will constantly need to refine the output after each algorithm or step in the process.
This is because each of the algorithms in the rule engine, although expressed as a set of tangible instructions, is dynamic in nature. The tangible instructions depend on many additional elements such as business goals, available resources, and the needs of the hour.
I wasted many hours doing it. I recommend to approach this carefully.
3. Evolution of thoughts
All of the files in your hard and cloud drive will never evolve by themselves. AI analyzes them in a certain way or expands on them in a certain direction, but it doesn’t enhance anything.
This is extremely crucial in content creation, especially for marketers. As you continually write blog posts or social media content for a particular topic, the quality of your content improves because you perceive the particular topic with more breadth and depth.
Over time, we accrue knowledge and use that to become wise. Think of the process as the churning of milk to produce cream. Knowledge (or truthful and objective information coupled with logical reasoning), with the power of our minds, turns into wisdom.
AI will never do that. That’s just facts.
Wrapping up: AGI is a mythical technology
The AI companies of the world are selling the dream of a machine that can think like humans only to milk more money from their investors. The CEOs and engineers in these tech companies are smart enough to know they are chasing the ghost.
If you think about it, the Gen AI models are already good enough. Now, the differentiator is how you use it. It is about understanding how LLMs work, the limitations of AI, and the implementation strategies.
And I have written this book to contribute in that direction. You should too.
—
Next Chapter: Coming soon
Previous Chapter: ChatGPT Is Best for Content Writing (4 Reasons)
Index (with Prologue): Content Writing With AI: Principles, Strategies, and Prompts
תגובות