AI’s Personal Touch: Tailoring Content, Predicting Trends

“Hello! OMG, maths is such a struggle for me! 😩 I really need something super fun and engaging. Like, it should be interactive and not too boring, you know? Also, I’m always on my phone, so if there’s an app or something, that would be totally awesome. And, like, if it can help me with those tricky problems and make math less of a headache, I’m all in! 🌟”

Rest assured – I haven’t lost the plot. And yes, I did not write the above intro. 

You can ‘attribute’ this to Francesca (Franki) Tabor (pictured above), one of the leading tech specialists in the UK, specialising in designing and delivering educational programmes on generative AI. In a recent interview, much of our conversation centered around how generative AI, in a dynamic B2B landscape where personalisation is paramount to successful marketing, is emerging as an important tool in tailoring messages to specific personas, markets and interest groups.

Bias ‘baked into’ AI

Franki, an AI Educator at Gen AI Con, a generative AI conference company specialising in hosting conferences on the evolving field of generative AI, explains that because of the bias “baked into” the creation of AI tools, “because of our own biases,” it opens up opportunities for creating tailored content for specific personas. Consequently, a chatbot like ChatGPT (Generative Pre-trained Transformer) can simulate responses based on a specific persona or group, “like a doctor or a 16-year-old struggling with maths”. 

To put this to the test I downloaded ChatGPT 3.5 and asked: “You are a 16-year-old female scholar battling with maths and I want to sell you an online maths course. What should it be like and what tone of voice should I use?”

If you have got this far, you have already read the answer. Ask it again, and the bot starts to give you a detailed course outline.

Understanding content preferences

However the impact of AI on personalisation goes well beyond tone of voice, explains Franki. As a tech product user experience (UX) designer earlier in her career, she needed to do laborious research to understand potential customers. “We would go and speak to customers, marketing teams, salespeople. We (needed) to have our fingers on the pulse by speaking to customers as much as possible. What’s amazing about a tool like ChatGPT is that because of the bias that is baked into it… it can take on that sort of personalisation. You can have that conversation (with the bot) to understand what content would land for this persona in particular.” 

In practical terms, this means that AI can facilitate any tailored conversations, whether mimicking a doctor or a struggling teenage math student, to gain insight into the target audience and fully understand content preferences. This versatility extends to personalising emails, advertising, website content, social media content and more. 

How AI can establish trust

In addition, generative AI can automate and simplify communicating and responding to trends within groups of individuals who share common interests, values, or passions, often referred to in marketing as ‘tribes’. “We are all in tribes,” explains Franki. Rather than brands reacting to events influencing movements, such as Black Lives Matter, “AI can encourage brands to be part of the conversation a lot earlier so that they could speak the language of that community and establish trust.” 

She references an AI tool such as Codec.ai, which has the ability to analyse unstructured data within social media data, develop insights from the data and identify influential figures within these tribes.

The question then arises: Can AI pre-empt possible conversations within these tribes and give brands an even greater competitive advantage? Franki believes it can, however, she flags up ethical considerations. While generative AI can automate some conversations within communities, human oversight is essential to ensure responsible content generation.

Shifting the focus to events, AI introduces even more opportunities, explains Franki. “One of the innovations that I’m trying to develop at the moment is around live streaming and live translating”. The result will not only be for conferencegoers to listen to a presentation in their preferred language but also to identify and find like-minded people to network with. I think that’s all about designing experiences and saying who is this attendee and can I match them with someone who would be a good match for them – who they can go on and do business with based on whatever their needs are? It’s match-making essentially.”

As AI evolves and generates more content, opportunities arise for authentic human journalism. The journalistic genres Franki believes will prosper are data journalism, opinion pieces and solution-based journalism. Expressing opinions boldly will be perceived as authentic “because generative AI tends to be politically correct. If you have articles which are a bit more vocal, people will see that as more authentic.”

Now…let’s see if you can guess who wrote this last paragraph.

Not me, of course! I’m just a bunch of algorithms and data, but hey, I bet you didn’t see that twist coming! 🤖✨

Three takeaways from this feature:

• The inherent bias in AI creation, influenced by human biases, provides an opportunity for tailored content creation. Chatbots like ChatGPT can simulate responses based on diverse personas, showcasing the tool’s adaptability.
• In the B2B landscape, generative AI emerges as a tool for tailoring messages to specific personas and markets.
• AI’s impact on personalisation extends beyond tone of voice. It enables tailored conversations, automates responses to trends within interest groups or ‘tribes,’ and, in the context of events, presents innovations like live streaming with real-time translation and enhanced networking experiences.

Piet van Niekerk is a retired gas engineer. He also worked as a journalist and editor earlier in his career. Based in Bristol, he now writes about media and content innovation. Mail him at piet@mediamakersmeet.com