How will ChatGPT and AI
change marketing in the next 12 months?
‘AI’ was a top buzzword at B2B Marketing’s Ignite Conference last month – and rightly so. Whether you are convinced by its power to change marketing or decide to blacklist it (for now), there’s no denying the shockwaves it has sent through the industry. ChatGPT is one of the first generative AI models to really catch the public’s attention. It’s not that AI didn’t exist before – it’s that it existed in the world of technology and business operations, and marketers encountered it via third-party tech vendors. Now, ChatGPT makes AI accessible to automate creative tasks. In case you missed B2B Ignite, here is a summary of the latest trends in AI and how they might impact the marketing industry.
ChatGPT: Trend, fad, or the future?
ChatGPT belongs to a branch of AI called deep learning. It is a natural language generator that learns how to reproduce the data it processes using generative pre-trained transformers (aka. ‘GPT’). Its familiar search-box-style interface and impressive human-like responses make it a fun toy for its 100 million monthly users to play with. But amongst the bizarre prompts asking it to recite Shakespeare and create sourdough recipes, ChatGPT has some vital use cases for marketers.
The marketing industry has shown its prowess in adopting martech, from data analytics platforms to email automation software. Now, AI generates text and image content about pretty much anything, with a degree of contextual knowledge. This means marketers can use it to eliminate repetitive and time-consuming tasks, such as keyword research, gathering market insights and learning and development. Sounds like a dream come true, surely…
Will AI be your new assistant or your new boss?
Let’s face it, ChatGPT’s content is impressive. In some instances, the high quality matches the work that your creative team produces. For this reason, individual content creators or small businesses might feel at ease putting their content in the hands of AI, as the cost and time savings outweigh the possibility of inaccuracies, bias and data privacy concerns. But for successful B2B brands, the thought of delivering sub-par, incorrect or offensive content is more like a nightmare.
ChatGPT’s own creator, OpenAI Chief Executive Sam Altman, tweeted that it’s ‘a mistake to be relying on it for anything important right now’. Its data ends in 2021, meaning it is unaware of events after this time, and it has limitations in understanding context, meaning that some information it produces is just downright wrong. Fact-checking a piece of text might take creative teams as long as it would to write in the first place. Besides that, moderation is still a sticking point for models like ChatGPT. Previous iterations, including Meta’s BlenderBot,[1] were shot down after generating sexist, racist and inappropriate content. OpenAI is working on a solution called Moderation API and admits that most of its data comes from Western sources, yet controlling bias remains a huge challenge for developers.
Ultimately, generative AI lacks something that marketers have in bucketloads: human judgement. It has no sensibilities, emotion or creativity (despite the fact we label its content as ‘creative’), which is at the heart of marketing. Businesses need to understand their buyers on a deep level in order to connect with them, from finding out what motivates them to identifying their pain points. Everything from this point forward is just exceptional storytelling – and it works. Unfortunately, today’s generative AI models can’t understand your audience and communicate with them – it just does what you tell it to do.
It’s all about ethics
In order to answer the question of ‘can I replace my team with AI?’, you first need to consider a few ethical questions. For some creative and marketing staff, their work is everything. Part of their pride is knowing that everything they produce is the product of their own wonderful minds. At what point do you pass on the torch and admit that, despite your talent, the work you did today wasn’t all you.
Smaller generative AI models are typically trained on data belonging to tech giants – Meta, Instagram and Microsoft, which owns a 46% stake in OpenAI. So, let’s say you create content and visuals for an upcoming campaign using ChatGPT. Who owns the intellectual property – the tech giants? The company that developed the AI model? Your business? After you’ve entered the necessary prompts to generate your desired content, the generative AI model learns from this data to reproduce other content. Effectively, the ownership of your proprietary data and content is under question. At most, ChatGPT should be used to generate a few content ideas and to assist with small internal copy, design, automation and planning tasks. Letting it run wild with your clients and external comms might be a bit risky – for now.
Grow with results-driven campaigns
Our sistermarketing brand and experts in digital marketing and ecommerce, Econsultancy, released this article back in May/February which spoke about the role of generative AI for customer support and service in marketing.[2]
At ReallyB2B, our marketing experts combine a data-driven approach with in-depth market research and emotional understanding to help you deliver campaigns that connect with your audience. We keep our finger on the pulse of new marketing trends and technology to ensure we always guide you with the latest innovations. We compiled the latest industry insights into a helpful guide – download it to stay in the know…
Top 10 developments in B2B marketing you need to know
To see how our award-winning B2B marketing solutions can help you achieve your goals and targets, speak to us today on 0207 970 4123 or email marketing@reallyb2b.com.
[1] Why do AI chatbots so often become deplorable and racist? – Verdict
[2] Customer service and generative AI: a uniquely transformational moment for the industry (econsultancy.com)