Author: Sebastian Scheplitz
It’s the talk of the town. It’s on the front page of every newspaper and every website. It’s in the middle of every conversation.
It’s called ChatGPT, and it’s highly touted as the next big thing.
If you’ve been following the content marketing industry for even just a cup of coffee – or even just “lived” on the internet for the past several months – you will have undoubtedly heard about ChatGPT.
After all, it’s all anyone and everyone and their mother has been talking about in the past several months. The advent of ChatGPT arrived to much fanfare – with reactions to its launch running the full gamut of positive, negative, and nonplussed emotions.
Some say it’s a disruptive, game-changing content marketing technology. Some say it’s the next evolution of AI that will make content marketers, writers, and SEO professionals obsolete.
Tech bros hype up ChatGPT as though it was the second coming of Jesus, while some outright dismiss the arrival of artificial intelligence in content marketing as just another fad. Hip today, gone tomorrow, so to speak.
But before we pen obituaries and plan funerals for the whole content marketing and SEO industry, as some might have, it’s only fair that we talk about the elephant in the room: mainly, the benefits of ChatGPT for content marketing and SEO, how to use ChatGPT for content creation, and more importantly, the undeniable disadvantages of ChatGPT for content writing.
What is ChatGPT?
In case you’ve been living under a rock, let us smarten you up about what ChatGPT is all about and why it’s been a lightning rod for hype and controversy in the content marketing industry.
ChatGPT is a chatbot powered by artificial intelligence built by OpenAI that’s been receiving plenty of coverage across mainstream and social media platforms. This category of generative AI-powered tech receives user prompts, takes them, then generates new content in response to the data set used to train the model.
What distinguishes ChatGPT from other AI-enabled chatbots is its capability to produce fluent, natural-sounding conversations that are nearly indistinguishable from those of humans. Its language model predicts the next word in a sentence using algorithms, allowing it to provide realistic-sounding responses to complex questions and inputs.
Suffice to say, ChatGPT is indeed a powerful tool with a vast range of potentially disruptive applications across industries, including content marketing.
Okay, the results are sort of impressive. Not bad at all.
But will it fundamentally change everything we know about content marketing for good? Is it the silver bullet that will make the entire gamut of content marketing professionals – creatives, writers, SEO wizards, and the lot – dare we say, obsolete?
Not with the current version.
At the very least, not yet.
Because, for all its innovative and disruptive potential, we must remember that the technology is literally in its embryonic stages. It’s highly likely to continue evolving and produce better results over time.
Skeptical as we are – after spending years on our craft to become great writers – we must concede that AI-powered technology will likely produce massive changes across creative industries, including communications and marketing.
It’s too soon to tell what those effects are.
Granted, there WILL be some individuals that will leverage such generative AI technologies in a way that undermines, if not outright harms, the creative process.
On the flip side, it’s just as likely that individuals will harness said technology to hone their craft and challenge themselves to create better content than the tool itself.
What Are the Potential Use Cases of ChatGPT 3.0 in Content Marketing?
Nevertheless, content marketing professionals need to be aware of how ChatGPT 3.0 can help with producing and curating content.
As it stands, ChatGPT’s natural language processing and GPT-3 technology enable it to create content that writers can tailor and spruce up to appeal to a particular audience, boost engagement, and ultimately direct traffic to their sites or socials.
We have to give credit to ChatGPT as a great tool for making basic queries and helping move along the creative process, especially for writers stuck in a rut.
But perhaps the most salient use case for ChatGPT for content marketing and creative teams is in its ability to help them research, curate, and find topics to write about – and contribute, rather than be the centerpiece of your content strategy.
Why shouldn’t you relegate your content strategy to ChatGPT?
That’s easy – and that’s because ChatGPT itself, like anything else, possesses inherent limitations. Its language processing model is based on language prediction to produce responses akin to humans – but just as humans are flawed, so can ChatGPT.
In fact, OpenAI itself doesn’t hide the fact that ChatGPT’s language processing model is NOT based on the facts of the content it produces. To wit, and to quote:
“ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers. Fixing this issue is challenging, as: (1) during RL training, there’s currently no source of truth; (2) training the model to be more cautious causes it to decline questions that it can answer correctly; and (3) supervised training misleads the model because the ideal answer depends on what the model knows, rather than what the human demonstrator knows.”
That’s why we here at Translation Royale would not use ChatGPT 3.0 to produce any sort of content for our clients – and we wear that badge loudly and proudly.
Here are five more reasons why we wouldn’t use ChatGPT 3.0.
1) ChatGPT 3.0 isn’t all that innovative or cutting-edge.
For all the hype about ChatGPT being a state-of-the-art, revolutionary tool, we should perhaps zoom out and examine its origins. OpenAI first introduced ChatGPT’s GPT-3 language processing model in 2020.
Prior to the emergence of GPT-3, there was GPT-2. GPT-2 worked in a similar fashion, but with one caveat: it was actually worse at trying to reproduce human-like responses.
ChatGPT’s GPT-3 model iterated on this with two new features: first, it could remember previous questions arising from the same conversation, and second, it offered a chat interface. OpenAI released this for the general public, and so began the craze which we hear of to this very day.
OpenAI is hardly the only generative AI stalwart making waves – Jasper, an AI copywriting app, has been around since 2020, and it can produce pretty compelling results (if you can afford its exorbitant prices).
Google, meanwhile, released its Bard chatbot several weeks ago, promoting a video peppered with factual errors.
That’s not a good look for any of these tools.
At all.
2) ChatGPT 3.0 isn’t as intelligent as you might have been led to believe.
As we’ve established earlier, artificial intelligence isn’t actually all THAT intelligent. In fact, all the errors AI has made over the years have added a word to our modern-day lexicon: bot-splaining. And ChatGPT isn’t above bot-splaining.
What is bot-splaining?
Let’s put it this way: that’s where the bot acts like David Brent or Michael Scott from either iteration of The Office – that is, where AI acts like that smug, self-assured moron who thinks he has an answer for everything – even when everyone can see that the emperor has no clothes, and just about as satisfied as someone being born on third base, thinking he hit a triple.
OpenAI, for its part, doesn’t hide the fact that ChatGPT is just as prone to bot-splaining when it says: “ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers.”
ChatGPT draws from over 300 billion words pulled from various corners of the internet. And it’s only as accurate as the text it was fed as input. Unless said input is actually up-to-date, ChatGPT wouldn’t know jack about current events, laws, or rules.
And, as we all know, you can take anything and everything on the internet as gospel truth.
Data can be biased, facts can be fabricated, and statistics can be fudged. ChatGPT can chop and splice these inputs together that might appear authoritative and factually correct – just as much as the Weekly World News is. Garbage in, garbage out.
3) ChatGPT 3.0 is lame.
There are certain prescriptive standards associated with good writing – and standards for every type of copy.
Case in point: elementary school writing would have you NOT split infinitives, use double negatives, or write essays without an introduction, a cohesive set of points, and a conclusion. You wouldn’t start sentences with particular words.
But the truth is that none of those rules are actually real.
Yet just like artificial intelligence, we learned to write using such standards. And when AI creates copy using such prescriptive standards – just like when we do – we all get very rote, rudimentary, and just downright lame copy.
ChatGPT’s output is proof positive of this.
Its long-form copy is incredibly boring and formulaic. Its long-winded answers are as interesting as watching paint dry. There’s not a hint of humor, style, or flair in its answers, nor does it use figures of speech like idioms, similes, and metaphors.
It’s completely humorless, not to mention the fact that it’s completely oblivious to colloquialisms like TL;DR or QRD. It overuses “however” as a sentence starter.
Again, take it from the makers of ChatGPT: “The model is often excessively verbose and overuses certain phrases.”
Brands that want that extra panache and flair in the tone of their writing – something inseparable from the clutter of content marketing – shouldn’t go all in on ChatGPT. See what we did there?
4) ChatGPT 3.0 is devoid of passion, emotion, or opinion.
ChatGPT is even lamer than your Trump-loving friend, because at least you know where your buddy stands. That’s all right, people can be wrong. But we digress. See, there are plenty of things, quirks, and idiosyncrasies that distinguish human from generative artificial intellect.
Humans are creative, drawing their creativity from personal experiences, subjective perceptions, and empathy. But perhaps nowhere is this clear distinction between human writing from ChatGPT output is that we draw from our opinions, passions, and emotions.
ChatGPT is aggravatingly neutral. Give it a prompt, and it will try to cover every angle without providing any real insight.
Chances are, you’ll get a bunch of equivocal responses beginning with “On the other hand…,” “In general,” No expert opinion, no in-depth analysis, no deep dive, not even a direct answer. No passion, no emotional connection, no nothing.
There’s just nothing in there.
And that emotional connection is vital to B2C and B2B SEO content writing.
Chances are, you will be familiar with Google’s E-E-A-T standards:
- Expertise
- Experience
- Authoritativeness
- Trustworthiness
We already know that ChatGPT fails on three counts – and that it can be as authoritative as David Brent.
Google wants content that adds value to the conversation. Opinions, essays, reviews, and comparisons – something that ChatGPT and other AI tools are a long way from bringing to the table.
You too, Jasper.
5) ChatGPT 3.0 can’t cite sources for crap.
B2B content marketing requires familiarity with legal reviews, just like large iGaming companies like most of our clients here at Translation Royale and our sister agency WeGotGame.
You can’t make unsubstantiated claims.
You can’t just fabricate statistics without citing where you pulled them from. Hell, you’re not even allowed to use certain words or make certain statements, like “Slot XYZ is fun and cute” (which we’re obviously not using anywhere).
The rules surrounding legal reviews vary wildly depending on region, and the day ChatGPT recognizes and learns all of them will be the same day MTV plays music videos. That is to say, never.
The tech behind ChatGPT is fascinating, even cool, within the context of technology. That’s why it’s so popular.
But not so for complying with the legalities. ChatGPT can spew out more fake links and non-existent sources than you can shake a stick at – but it can’t actually define the sources from whence it takes its information.
If you’re relying on ChatGPT for your iGaming content marketing, we hope you have a nice, fat budget for a professional fact-checker. That’s all we’re going to say.
6) ChatGPT 3.0 can be helpful…sometimes.
All right, all right, we’ll admit it. Even in its nascent and undoubtedly fallible state, ChatGPT can actually help content writers with some of its features.
For instance, ChatGPT can produce oodles and oodles of short-form content in mere seconds, such as headlines, subject lines, and social media posts.
In fact, ChatGPT is a helpful tool to help content writers come up with topics, keywords, and outlines that they can then take to the next level and fine-tune with that divine spark of humanity we all so desire from our copy.
Ultimately, some of its key features can save hours of wasted time and bother over the long term.
But while ChatGPT has the undoubted potential to enhance marketing efficiency, content marketers should only ever use it and other similar AI tools as a supplementary tool to assist humans and help them achieve their goals – not replace them entirely.
ChatGPT could provide creative content and support content ideation. However, the human factor is still essential for examining outputs and creating marketing messages that are consistent with a firm’s business strategy and vision.
A business that does not have a strong marketing strategy before integrating ChatGPT remains at a competitive disadvantage. However, with appropriate marketing strategies and plans, ChatGPT could effectively enhance and support existing marketing processes.
7) ChatGPT 3.0 will change the way we create content – and that’s a GOOD thing.
Let’s face it: the only content writers who should panic about the emergence of ChatGPT are the ones with questionable skills.
That’s because the only content that AI can possibly replace at this very moment is the bland, tired, mediocre cookie-cutter content that content houses churn out on an assembly line to build web presence and drive some traffic.
The fact is, there will always be a market for such hackneyed content because of how cheap companies can get it.
We get it.
Paradoxically, that same market reliant on mediocrity is the same one driving the market for high-quality, emotionally-charged, and compelling content that connects to audiences in every part of the world.
There will always be a demand for writers that are fluent and nimble enough to hit that sweet spot where they nail the appropriate tone and voice required to create truly outstanding content.
Experts with opinions, fluency, and flair will quickly become highly sought-after superstars. And behind every great piece of writing will be the fact-checkers and editors – the true beneficiaries and key players who will be at the forefront of the ChatGPT paradigm shift.
This is the kind of engaging, compelling content that only a living, breathing human can write.
And that’s exactly what we’re all about at Translation Royale. Our entire goal as a human-focused, professional translation, content marketing, and localization agency is to produce results that speak volumes for our iGaming clients and their audience.
The proof is in the pudding – can you taste the difference yet?
Header Image Source: frimufilms on freepik
How did you like Sebastian Scheplitz’s blog post “Who’s Afraid of Big, Bad ChatGPT 3.0? Here are 7 Reasons Why Content Marketers Shouldn’t”? Let us know in the comments if you have anything to add, have another content idea for iGaming blog posts, or just want to say “hello.” 🙂
1 Comment