Can AI-generated content be original? How can you use content-generating AI technologies like ChatGPT effectively? Read on to find out.
In 2008, Gary Vaynerchuk famously extended Gates' quote to "Content is King, but marketing is Queen and the Queen runs the household". Of course he was referring to digital marketing channel context and how the content needed to be developed not only for the channel/medium but also tailored for personas, email lists and social communities. Content needed to be unique and engaging enough to captivate your audience’s attention and elevate your brand awareness. The new buzzwords became personalization, multichannel and omnichannel. The more we thought we could understand our customers, the more we believed we could tailor our content to deliver the right message to the right channel, and indeed the individual.
But the more personalized content became, the more we started losing customers because they found the level of personalization just too creepy. It turned customers off brands, instead of increasing their loyalty. As it turns out, most customers are happy for messages and offers to be tailored for them, and for their brand to be helpful and provide useful information, but at some point, personalization definitely crossed the line from “friendly helper” to “stalker” behaviour. And repeating the same personalized content uniformly across all digital channels just jacked up the creep factor to an intolerable level. Who hasn’t wondered (either seriously or in jest) whether Google was listening into their face-to-face conversations? As an industry, we just haven’t been able to get the balance right.
Authenticity and trust are key to building a brand community
Enter the era of authenticity, data privacy and respect. We’ve cut out the “marketing BS” and are speaking to the audience openly from the heart about things that really matter to them instead of selling our product features and services. Things that show our audience who we are as people, as a brand, and as a community. Things that show them what we stand for, what makes us laugh or cry, and things that encourage them to become a part of the brand’s activities, and contribute to them. By treating them with respect and nurturing, not manipulating, them along their customer journey, we started to empower them to make the best decision for them. This approach aims to establish a mutually beneficial relationship for both vendor and customer, as the customers become staunch supporters of their favourite brand largely due to the fact that they are a contributing and respected member of that brand community.
Content composition therefore needs to not only take into account the target audience, the channel, the context, the brand archetype, and the trends within the market, it also needs to be authentic, respectful and personally engaging. It needs to be interesting enough to attract and retain the attention of its audience and encite them to join and participate within the brand community – generating their own brand content. It needs to build a relationship of trust between the brand and the consumer. This is the goal of most marketing teams. If your community members are spreading your message, sharing your values and joining or even leading the global fun and conversation, then the content being generated is as authentic as it can be, and can globalize your message in a heartbeat if the message resonates.
Truly great authentic content doesn’t just happen by magic. It has to be triggered. The seed you plant must spark the right kind of creativity in the minds of others.
Working out what those seeds are is a seemingly never-ending process of research and ideation. THIS is the less glamorous side of composing great content – it can be a bit of a love-hate relationship.
We love great content. We are passionate about creating it. We are totally mad about it. But some mornings we just can’t face that blinking cursor (pun intended) – mocking us for staring at the screen with our minds blank. Composing quality authentic content takes time and effort. Being constantly creative can be exhausting. Add to this the scarcity of great content composers, it’s little wonder we have started looking at ways we can outsource routine tasks and focus our real resources on the type of content that drives community engagement, brand awareness and ROI.
Can AI-generated content be original?
AI, of course, exists today to help content writers improve the fruits of their labor. If we think about Grammarly, or sentiment analysis tools that can provide marketers with feedback on the style and tone of their customers’ comments (like we use in Kentico Xperience 13), human-generated content is already being checked and enhanced by AI – but the content must be written first.
November 2022 tipped the world of content composition on its head and made everyone take note. Suddenly, natural-sounding, highly plausible ad and website copy, social media posts, blog articles and indeed any kind of content could be generated by anyone simply by entering a few well chosen keywords into ChatGPT or similar generative AI programs.
Some people were elated! Some probably wanted to despair. Microsoft shareholders were no doubt gleeful watching the sharp increase in OpenAI’s market value.
But no matter where you were on the spectrum of diverse reactions to generative-AI, you have to agree that this technology has been the focus of many, many conversations in recent months.
When commenting on the recent ChatGPT technology release, Bernard Mar argued that everything it writes or creates is based on something that has been written before. He writes “This means it isn’t actually capable of original thought or creativity in the same way as humans.” (Forbes, January 17, 2023)
Many AI researchers have tended to agree with this line of thinking, arguing that while these technologies essentially process and combine previously established data in novel ways, their learning is constrained by some initial principles provided by humans, and therefore lacks the creativity and originality of human mind processes to make associations beyond these constraints.
Whilst I’m sure there are many people who would love to agree with this thinking, Mark Twain, should he be alive today, would probably disagree with them. In 1933, he shared his thoughts that: “…substantially all ideas are second-hand, consciously and unconsciously drawn from a million outside sources, and daily used by the garnerer with a pride and satisfaction born of the superstition that (s/)he originated them.” (Letter to Anne Macy. Reprinted in Anne Sullivan Macy, The Story behind Helen Keller, 1933, p162)
To Mark Twain therefore, each thought is not original, but rather a collection of experiences and inputs gathered and synthesized over the course of our lives. Moreover, Mark Twain was also famous for his writing on the unconscious and conscious appropriation of the content of others. He believed that if you drew together and repeated content or ideas from the depths of your mind without knowingly copying it from someone else, “mistaking it for a new birth instead of a mummy”, then this was not plagiarism. Knowingly stealing the ideas made you, well, by both his and today’s legal standpoint, a thief.
Kirby Ferguson, Canadian filmmaker, must have been inspired by Mr Twain, waging war against plagiarism laws and promoting the remix culture. In a now quite famous TEDTalk, Ferguson stated “Nothing is original, everything is a remix!”, emphasising too that we are all remixing ideas that were previously shared and drawn from others. Others following this culture cite famous self-proclaimed unoriginal artists and inventors like Mark Twain, Coco Chanel, Henry Ford and Pablo Picasso as being likeminded individuals despite being held up to the world as geniuses.
From this point of view, we are all influencing one another all the time.
So if we agree with this premise of being a composite of human shared history and our own lived experiences, if we are indeed “standing on the shoulders of giants”, then how does this notion translate to authentic content? Can content produced by generative-AI be still considered “authentic”?
Generative AI versus “the authentic me”.
We are embracing AI, and we will continue to do so, but with caution.
We advertise that our social posts were AI-generated or conversely were written by “the authentic me”. We feel the need to publish the query we entered into the software to prove the keywords we entered and what the software’s response was. Universities and schools have scrambled to exchange digital assessment processes to in-person verbal exams or good ol‘ pen and paper tests. The notions of trust, creativity and other human traits like humour or bias are becoming increasingly central in our discussions to distinguish the human benefit over AI-generated content.
Recent history has encouraged many of us to tread carefully and not explicitly trust AI just yet. We are aware of the previous negative biases learned by its predecessors. We have been advised that AI-generated content might very well sound plausible, natural and convincing, it can still be factually incorrect. It might not even respect internet content, attribution, copyright and data privacy laws.
In the new world where the market now assumes or is suspicious of the content that marketing teams are creating, how do we build real relationships with them and our community? How do we maintain a relationship of trust? Perhaps it’s time to take a fresh look at what we mean by authentic content and think of it in terms of honesty, personality, respect and attribution. Content writers and digital marketers need to ensure their brand content remains relatable and full of their specific tone or kind of humour that their community knows and loves about their brand and voice. They also need to take great care to ensure their brand remains trustworthy in the eyes of their community and that it continues to uphold its reputation.
Use content-generating AI technologies effectively
So while we agree that content-generating AI can definitely help us by generating plausible content that saves us a lot of time, it has not yet reached the point where marketers can blindly rely on it to do a content writer’s job. It can definitely help you to focus your resources on what really matters – that authenticity part. So here are a few commonsense tips for using content-generating AI technologies effectively:
- Use it to kick off your research - summarize the market, analyse a competitor’s web presence, or gather ideas on a specific topic.
- Leverage it for low-hanging fruit or routine tasks – for example, preparing summaries of existing long-form content, summarizing long-form content for its promotion across different marketing channels (like social, email newsletters and the like), writing lower-value articles, posts, or SEO descriptions and keywords, and preparing video subtitles.
- Always include data sources in your requests for longer texts, and verify the information provided from original sources before relying on it.
- Revise and tailor the voice, humour and terminology to ensure it is written in your brand voice – editing is faster than creating from scratch.
- Implement publishing workflows – Human-review the content against your publishing standards before allowing it to be released to the public. That way you will catch and correct it when it inadvertently spells your product name wrong! (true story).
Want more tips like these? Check out our resources.