Artificial Intelligence is no longer a buzzword. Thanks to the recent breakthroughs in recent years, AI has become an essential part of our lives, touching nearly every aspect of technology. From Siri and Alexa, Facial recognition to AI CEO’s, AI is everywhere. We are now seeing AI come into content creation, with algorithms that can write books, create content, and even make music.
However, Artificial Intelligence is much more than just facial recognition, social media algorithms, self driving cars, and Bach imitations. Businesses are using AI and machine learning to analyze huge amounts of data to make sure they are writing relevant content in the right tone of voice, for the right audience, at scale.
AI will inevitably become an integral part of digital content creation. But despite huge progression with AI, it is not going to replace humans. In this article, we will explain why AI will assist rather than replace content creators, as well as the role both have to play in the future of content war.
Before we explore the reasons why AI cannot replace writers, we need to understand how computers translate written texts and spoken languages. The root of this is NLP (Natural Language Processing) which is a branch of AI that makes this possible.
The rise of AI
“A computer would deserve to be called intelligent if it could deceive a human into believing that it was human”
– Alan Turing – The Grandaddy of Modern Computer Science
Alan Turing is considered one of the fathers of computer science and AI. He was part of the team of scientists from Bletchley Park who cracked the German enigma code, ultimately leading to the end of World War II, and victory for the allied forces. No small feat.
The results of his work were used to create Colossus, the world’s first programmable digital computer, and then in the 1950’s, he created the Turing Test which is designed to measure a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from that of a human. The test is conducted in a written conversation between a human, a human judge, and a machine. If the judge cannot tell which one is the machine, the machine is said to have passed the test, and genuine Artificial Intelligence will have been achieved.
To date, no machine has passed the Turing Test.
But AI has come a long way since the 1950’s. From the 2010 onwards, a much more powerful version of AI has been created, based on Big Data, statistics, deep learning, and neural networks.
This AI forms the basis of NPL.
Natural Language Progression.
Natural Language Progression (NLP) is a technique that employs both Natural Language Generation (NLG) and Natural Language Understanding (NLU) to extract meaning from what is said, and then to generate text and speech respectively.
NLG and NLU have fundamental differences.
- NLG is the ability of the computer to generate natural language text or speech.
- NLU is the ability of the computer to understand natural language text or speech.
When one of these techniques is used alone, it can be considered as a subset of Natural Language Progression.
GPT-3 by OpenAI is the most powerful AI machine learning tool built to date, that generates text based on what it has learned from reading approximately 10% of the internet, and features about 175 billion trainable parameters (the values that a neural network tries to optimise during training). That’s over 10x bigger than GPT-2.
This inevitably brings up the question: if GPT-3 language models are trained by reading the internet, how can it produce original thoughts? The simple answer is that it can’t. Rather it replicates the content found in natural language, all over the internet.
The algorithm learns the way that humans use language, which allows it to create sentences that are semantically correct and stylistically similar to what is found in natural language.
But it’s important to note that it doesn’t take from one source, but rather an aggregate of all of the sources available to it to build a broad picture of that topic. Then, taking prompts from the user, it can take input text as an input and transform it into what it predicts the most useful result will be – building content word by word.
This process ensures content is not duplicated – but recreated. Leaving the author as the subject matter expert to adjust copy to their tone and match their point of view.
Human and machine relationship
With the advance of AI, humans and machines are likely to have a much closer relationship in the next few decades. This will bring (and has already brought) us more convenience, but with this come more challenges. The most important challenge is AI being the cause many job losses. This is particularly felt across industries like manufacturing, but far less so in more creative fields.
Even RoboCob knew this.
This is because, as the Turning Test (and to a lesser extent RoboCob) continues to prove, there will always be a need for input human from beings to cover areas that AI cannot.
Nuance, emotion, humour, objectivity, subjectivity, love, anger; there’s a long list of human characteristics that AI can attempt, but is as yet unable to perfect.
This is why AI will always need a human touch.
AI will improve – it’s getting better and more intelligent everyday. However, we are long long way from AI completely taking over human creativity.
How AI writing is changing the way we create content
Having said this there are a lot of areas (as is the case with other industries) where AI is being used for some of the heavy lifting.
As you know, the problem a lot of the time is consistency – Consistency in brand messaging is one of the many factors that determine the growth and success of your business. Big brands know how to portray their brand consistently and at every customer touchpoint. Your content needs to carry a unique and identifiable voice, style, and pitch across all distribution channels.
Consistency here establishes your credibility, builds trust, and strengthens your reputation. And the more consistent, high-quality content you produce, the better it’ll be for your SEO efforts.
Not an easy task to do at scale. Time constraints, staff turnover, content outsourcing – there’s a myriad of reasons why a consistent tone of voice isn’t easy. Which is a major problem if you want to keep your audience engaged.
Training AI helps overcome the consistency question. Because it can be trained to learn your companies tone of voice, AI can replicate your tone of voice and your companies values when generating content.
From here, content creators typically tweak the content to make sure it matches your tone of voice exactly, (though you can change the tone of voice with the click of a button as well), then sending it off to collaborate on if needed, or sending it off into the ether for the internet to get their hungry little mitts on.
When it comes to SEO, AI is changing the way content is created. But it is not enough to just create great content, post, and cross your fingers that the right audience will see it. The importance of optimzing your content so that it ranks well in search engines results can not be over stated.
AI can help you out here. Instead of manually creating content to rank well in search results, content creators are now able to make use of sophisticated algorithms to create content that is optimized for search. This way, your blog and website are primed for searchability, so it is easy for consumers to find what you have to offer.
With the rise of AI, it’s no wonder that content creators have started to use this technology to help them write content faster. With algorithms taking care some of the research (GPT-3 has read over 10% of the internet remember), keyword finding, idea creation, and building out first drafts of blog articles in 5 minutes, you can spend more time on improving your content and adding in the secret herbs and spices that make your writing yours.
These are the areas where AI writers do best. Assisting the pro’s with the menial tasks, so that they can spend more time being creative, and hit their content creation goals faster.
AI’s role in the future of writing
Although artificial intelligence will not fully replace writing as a profession, it will replace certain writing jobs. Writers that do not adjust to this will likely find it difficult to keep up, but those who will adjust will continue to thrive.
Feeding the content beast
The hunger for creating more content is seemingly never ending. And when content marketing costs 62% less than traditional marketing while generating 3 times as many leads, you can understand why it’s likely that this gluttony for content is here to stay.
This then begs the question – how do writers create engaging, compelling content at scale, without compromising on quality or consistency?
You probably know where I’m going with this one.
Because AI can help out with the heavy lifting of idea creation, research, generating subject headlines, create article outlines, meta descriptions, ads, social media posts, etc, etc; content creators are able to make first drafts in minutes. So your can spend your time as the subject matter expert to make the content your own 10x faster.
Wearing many literary hats
Part of the content beast equation is making sure we are writing in the right way for the right audience. Tone of voice is vital for this.
As mentioned earlier, because AI can be trained to learn your companies tone of voice, AI can replicate your tone of voice and your companies values when generating content, even if you have multiple personas in your target audience. This makes AI not one writing assistant, but many. Switching tone/persona to ensure that your message lands correctly.
Assisting non-native speakers
There are billions that want to learn English, but learning the nuances of the language – grammar, spelling, syntax, sentence structure, etc do not come easy. Reading this study, early results indicate that AI tools will be useful for future English language learners who need more structured assistance than traditional word processors.
Why AI isn’t replacing humans
AI is not conscious
As we’ve seen, AI can do incredible things – but creating content that really truly resonates with people is not one of them. This is because content is created based on statistics and probability, not generated from a conscious mind. Input from a human is still required to tell the AI what to write about, in the same way that a human needs to tell a calculator what equation to solve. Because neither are able to think for themselves.
Both are examples of tools to assist humans – not replace.
Humans love magic
But beyond the obvious, humans will always want to read what a human has written, not what a robot has generated. Sorry Mark.
We want to hear and understand the point of view of the writer – often whether we agree with what they have to say or not. The tone, emotion, rhythm of the story – these are the magical parts that we want to read, and hear, and feel.
But these are the parts of humans that AI is unable to replicate as yet.
Even though around 75% of this article has been written by AI, there are parts that needed to be written by a human to make it more enjoyable (I hope) to read.
AI works best when assisting content creators – not replacing them. This new world, this co-creation utopia, content creators are going to spend more time on thought leadership content, conducting in depth research, carving out their narrative, and the creative side of writing. The role of AI will be the heavy lifting behind the scenes. Helping out with ideation, outlines, and getting rid of the blank canvas.