AI-Generated Content – The Good, The Bad and The Ugly

AI-Generated Content – The Good, The Bad and The Ugly

“In a kingdom ruled by a curious king, a magical machine called “AI” was discovered. It could generate all sorts of content, from articles to music and stories. The king ordered his jester to test it out, and the machine produced a funny story. As word spread, more people used it to create catchy ads, compose new songs, and invent recipes. The kingdom became known for its incredible AI-generated content, bringing joy to all.”

Our ChatGPT request: “Write a short introduction about AI-generated content in a form of a quirky story.”

“In a kingdom ruled by a curious king, a magical machine called “AI” was discovered. It could generate all sorts of content, from articles to music and stories. The king ordered his jester to test it out, and the machine produced a funny story. As word spread, more people used it to create catchy ads, compose new songs, and invent recipes. The kingdom became known for its incredible AI-generated content, bringing joy to all.”

Enjoying this short, quirky story written by AI-generator, ChatGPT? How well does it hit the brief we gave it? There was no input or modification from our end – this text is exactly as we got it in its raw form.

The rest of this article is written by a real human, we promise! Read on to learn more about the good, the bad, and the ugly of AI-generated content.

 

What’s AI-generated content? 🤔

AI-generated can be most simply defined as content that is written by a machine instead of a human. A person, like you or me, enters a text request (in this scenario, our brief for a short story) that the machine then processes and creates content based on this information.

You may be surprised to learn AI-generated content has existed more than seven decades already.

The earliest iteration was created in 1951 by Christopher Strachey, who later become Director of the Programming Research Group at the University of Oxford.

We often think about AI as a fairly new invention, but it’s important to note here that it’s been around to many years in different shapes and forms, whether that’s for medicine, phone technology, chatbots, SaaS, and more.

It’s only in recent months that AI for written content has been so widely available to the public, and the resulting AI-takeover of the internet ensued.

In an article exploring how Americans see AI advancements within their daily life, around 33% of the interviewees knew about AI’s capability to writing news content, with only 16% thinking it’s a major advancement compared to 45% who don’t.

 

So, why is it only just becoming a hot topic? 💥

You’re probably wondering two things:

  1. Why has AI-generated content only recently become available to the masses?
  2. Why has it only just gone viral?

 

Before, AI was open to the public, but was generating some pretty controversial stuff. Think self-harm and terrorism – and that’s the tamer end of the spectrum! Even recently, BingAI threatened to destroy everything – yikes.

So, what’s so revolutionary about applications like ChatGPT? They seem to be a lot more controlled and better defined, so it’s more suitable for everyday use across the world.

ChatGPT is King – for now…

We couldn’t write an AI-generated content article wouldn’t go without mentioning the elephant in the room – ChatGPT.

Launched in November 2022 and created by OpenAI developers, ChatGPT is a free AI tool that rose to international fame essentially overnight.

If you’ve been on LinkedIn in the last four months, chances are your feed is full of tips, tricks, and more about this nifty little tool.

With a sharp rise in popularity bringing an influx of free users to their platform, OpenAI introduced a new paid-for portal.

The level of consistency and detailed queries it can handle make this platform a favourite for most users.

 

Is ChatGPT the only option in the AI landscape? 🕵️

No. There are others available to the public (free or paid) that can be used for different purposes. It just so happens that ChatGPT doesn’t specialise and will answer most queries a human needs to process.

  • Bing and Microsoft:
    https://www.bing.com/new
    In an effort to stay ahead of the competition, Microsoft developed a new AI-generated feature on their search engine, Bing. They encourage users to use the chat section to ask questions as if they were talking to a real human. BingAI is based on the integration with OpenAI (aka ChatGPT), so may return similar answers.
  • Brave: Summarizer
    https://brave.com/ai-summarizer/
    Brave Search recently introduced the Summarizer on its results pages, an AI tool that summarises web search results in a quick and to-the-point answer: “This produces a more concise, accurate answer, expressed in coherent language,” Brave say on their blog.
  • DuckDuckGo: DuckAssist
    https://spreadprivacy.com/duckassist-launch/
    DuckDuckGo developed DuckAssist, a new feature that generates natural language answers to search queries using Wikipedia. Still in Beta, it appears that DuckDuckGo is looking to widen the base source this ‘Instant Answer’ generates.

 

Ai generated content explanatory infographic by Furthermore Marketing

 

 

Advantages and limitations 👾

Although very new on the market and free for public usage, here are the main advantages that AI-generated content offers:

  • Quick to use,
  • Good for brainstorming,
  • Answers we can define and drilldown,
  • Fun to play with.

 

It’s amazing to see your answers generated right before your eyes without the need for looking it up through multitude articles. This tool will remove most repetitive tasks while potentially reducing errors.

As mentioned above, the main advantages with AI content are to assist the human with generating ideas, facts, and content structures to better understand the topic at hand.

But there are some limitations that we must be aware of due to the machine still being in development…

  • Grey area with plagiarism and generic content
  • Your Money or Your Life, aka YMYL, generated content is not monitored
  • Everyone can get the same results
  • Generated content with ChatGPT is not up to date

 

If your topic touches human wellbeing or variations of life-based topics, you may find ChatGPT’s answers very limited or not accurate enough.
This is where its limitations remain (for now), which not all users are aware of!

It is important to understand that generated content must be read by a human and verified for its accuracy – a scandal and fact that has already been in the media.

Bear in mind as well that the answers you get could be similar to others’, running the risk of making your content generic and unoriginal.

Finally, if you open a new chat on ChatGPT, it does indicate that answers generated have “limited knowledge of world and events after 2021”. If your topic is about something very recent, you might not get the answers you’re after.

Intro Screen Warnings on ChatGPT

Here’s an example based on a request on March 2023:
“Where is the Queen of England?”

ChatGPT answers:
“As an AI language model, I do not have access to real-time information, but as of my knowledge cutoff in 2021, the Queen of England, Queen Elizabeth II, resided primarily at Buckingham Palace in London, England. However, she also has other royal residences such as Windsor Castle, Sandringham House, and Balmoral Castle where she spends time throughout the year.”

ChatGPT answers about Queen of England

As of 24th March 2023, OpenAI released new plugins that could bypass this issue that allow browsing websites and give more accurate, up-to-date information. More about this information here.

 

Will content be relevant and trustworthy over time? ✍🏻

As ChatGPT revolutionised the online content creation, people are naturally worried about how accurate or relevant online content will become.

Will it be trustworthy? Will it be true to the knowledge of the company or website you land on? Or will everyone be a professional in all topics thanks to ChatGPT?

This worry has people questioning AI-generated content, anticipating that a lot of companies are seeing a shortcut to content creation, using it without moderation or publishing facts that are not verified by a human.

Will Google find a way to restrict or penalise this content? Or will you have a tool to detect what is human versus what isn’t?

 

AI content detectors 🦾

Since ChatGPT slingshotted itself in our lives, other companies are seeing opportunities in a different way. You can now find online AI detectors that scan any piece of content and scores the probability of it being an AI written content or human.

So far, these new tools help combat this online trend and make sure that what you read is what you get.

However, new tools come into play by paraphrasing AI content and re-writing these into a more human language, fooling the detector and scoring the content as human.

So, is there a way to accurately fight online generated content? Or should we embrace the new “normal”?

 

Is it against Google ranking rules/SEO? 🚨

Fret not! To answer the questions above, Google is coming in force to help filter AI-generated content.

Introducing EEAT, the new framework to identify good content to rank high in the search results. EEAT stands for Expertise, Experience, Authority and Trust. Four elements that will help Google identify most relevant content.

Chances are, ChatGPT doesn’t have any personal experience about any given subject nor expertise. This could be a chance for search engines to filter down content and prioritise those that are written with a brain rather than code.

Take this excerpt from Google’s official FAQ:

Q: Should I use AI to generate content?
A: If you see AI as an essential way to help you produce content that is helpful and original, it might be useful to consider. If you see AI as an inexpensive, easy way to game search engine rankings, then no.”

👉 If you are still wondering the benefits of SEO content, maybe this quote should do it?
This is what professional SEO services look like.

 

What it means for “human” jobs, positions and teams 🏃

ChatGPT is not only scaring users and readers alike on the web, but also worrying copy and content creators where their livelihood relies on research and the art of putting words together.

Will ChatGPT take this away from them? Will employers see a shortcut with this new technology?

The impact of ChatGPT is still unclear, as is its future within our society. As it stands, this technology is more of an opportunity to reduce repetitive tasks and focus on the most important goal: producing high quality content.

With its current limitations, ChatGPT doesn’t rival a copy or content creator that can interview experienced people and elevate content with trustworthy sources.

As a marketing agency, our top tips to include ChatGPT within your existing team include:

  • Using ChatGPT for brainstorming
  • Gathering facts as a listicle for quick content gathering
  • Challenging the machine and finding new content opportunities

 

Don’t ignore AI-generated content,
Be smart with it! 🧠

Is Terminator’s future becoming our future?
Is ChatGPT going to replace the human brain for online content?

Although ChatGPT raised the bar with their new AI-generated content, there is still a lot we need to understand and experiment with. Its limitations are an opportunity for new job roles and new skills on the market.

But will we mere mortals know when the AI-content is accurate? Will we know how to detect its own content as years progresses? Google’s search engine framework EEAT offers a fighting chance to combat AI-generated content that will prioritise human content based on experience and expertise.

All in all, we should learn how to master the new tool and make it work for us, not the other way around!

If your team needs assistance about how to use AI-generated content efficiently, or your company is looking for extraordinary content to be (humanly) written, contact our team today! Contact us

Share this post