Marketers are polarized about the impact of artificial intelligence on the marketing discipline. We believe the positives of AI marketing will far outweigh the negatives. In 2024, AI will become a widely accepted tool for marketers.
AI marketing hallucinations
There’s no denying that we have seen some AI marketing failures of epic proportion recently.
What people saw on social media vs. What they saw when they turned up
An events company in Glasgow got carried away using generative AI to create fantastical images of a Willy Wonka factory, promising customers a magical in-person experience where “chocolate dreams become reality.” When people showed up, what they saw was pathetic—some lame decorations, and a sad actor pretending to be an Oompa Loompa in a deserted warehouse. Irate customers called the police, the disaster went viral, and social media has branded the event as “Scotland’s Frye Festival.” The worst part? No chocolate at the event!
Microsoft’s Bing chat service recently told a New York Times reporter that he would be better off leaving his wife. Air Canada’s chat bot erroneously promised a customer a refund, which the company refused to pay, but was ultimately forced to honor, following humiliation on social media and legal action.
The negatives of AI marketing are over-emphasized
Such high-profile incidents have fueled the skepticism towards AI marketing. Many of the issues are serious and the marketing profession will need to address them as it continues to embrace AI.
Biases inherent in the training data set for AI models could lead to discrimination in marketing. For example, Google’s GenAI app Gemini found itself in the crosshairs of the culture wars when it refused to generate images of white people, presenting only Black and Asian Australians, equated Hitler with Elon Musk, and labeled Indian Prime Minister Modi’s policies as “fascist.” The tech giant has apologized for some of these gaffes, and pledged to retrain its AI engines. The problem was that Google had inserted a diversity instruction to compensate for the fact that AI finds more pictures of white people on the web than other ethnicities. The issue of biases is a solvable one.
Marketers also lament that messaging created by AI marketing lacks the human touch and fails to engage as effectively as human-created communications. Further, differentiation is at risk since any brand can use AI. However, in a recent study, consumers preferred AI-generated content to human-generated by 54% vs 46%. AI marketing will continue to improve and generate messaging that is as engaging as human marketing. But when it comes to overall brand strategy and positioning, AI marketing will not replace human marketers. The human ability to surprise, delight, engage, be unique, think ahead of consumers, and find the next big thing, will always be essential.
Some fear that AI will put marketing jobs at risk. But surely that is not a reason to refuse to use it. Harnessing AI for marketing will require marketers to retrain themselves with unique skills such prompt engineering. And, if there are automated tasks that AI can take over, making marketing more efficient, and freeing up practitioners to focus on bringing more creativity into marketing, that can only be a good thing.
AI can be used in many ways for marketing
Skeptics of AI marketing tend to view the technology from the lens of its generative capabilities.
But GenAI, the ability to harness AI to generate text, images and video, is only one type of AI. There is also analytical AI, the ability to analyze consumers, predictive AI, the ability to identify future trends, and computational AI, the ability to find, process and learn from large sets of data incomparably faster than any human can.
While marketers are attracted by the generative aspects of AI marketing, it is the more fundamental machine learning capabilities of AI that will have a more enduring impact on what we do.
As these other capabilities continue to advance, AI will become an essential enabler for marketers—marketing’s Augmented Intelligence.
AI marketing has tremendous positive potential
AI will improve marketing in a number of ways, increasing its efficiency, enhancing its impact, and reducing its risks.
AI marketing will provide better personalization. Netflix, Amazon and Spotify are already deploying AI engines to understand individual consumer preferences and serve up new content to them. 70% of what people watch on Netflix and 35% of Amazon’s sales come from recommendations made to consumers by AI.
AI marketing will transform the customer experience. For example, the 24/7 available AI-powered chatbots of JPMorgan Chase are able to answer questions, assist with purchasing decisions, and provide recommendations. AI has boosted customer satisfaction by 15%, reducing wait times, and improving operational efficiency. The Walt Disney Company uses AI to analyze vast sets of data, such as wait times, purchases, sentiment, and uses this to optimize the park experience with better pricing strategies and crowd management.
AI marketing will help predict customer needs. For example, HSBC uses AI to analyze customers’ banking activities, demographics, employment, credit, to predict future desires from individual customers. This allows the bank to proactively offer personalized products such as loan offers or investments. Macy’s puts its AI engine to work to predict fashion trends, optimize inventories, tailor marketing and inform new product design.
AI will increase marketing efficiency in many different ways. Companies can leverage AI to automate many repetitive tasks, such as email marketing and lead nurturing. They can also use AI to conduct sentiment analysis or examine customer data to optimize the creation and targeting of ad campaigns. For example, Adidas uses AI to identify customer personas, and serve up the most relevant ads on social media, which has led to a 20% increase in click-through rates and an improved brand perception.
AI marketing will decrease marketing risks. By using AI in ad testing and attention prediction, companies will be able to conduct smart experimentation. They will learn what works and what doesn’t, and deploy only the most effective communications to the most relevant audiences.
And yes, AI marketing will help with content creation. The generative aspect of AI will continue to evolve and be used more and more by marketers and content providers. For example, the Associated Press uses AI to generate news alerts, personalize news feeds, and curate content. OpenAI’s recent launch of Sora, its multimodal AI text-to-video app which can generate high quality videos in response to a prompt, is already creating a stir within the marketing community. AI marketing, with its greater efficiency and ability to easily generate images and videos, will also help smaller companies to market on an equal footing with larger competitors.
How AI should be used in marketing
There may come a day when the ultimate goal of Artificial General Intelligence (AGI) is achieved, with machines as smart as and indistinguishable from humans. Until then, we should tap AI marketing for what it is best at doing, and not try to force it to do what it can’t. AI marketing can know, but it cannot yet be wise. It can have intellect, but not instinct. It can decide, but not judge. It can predict, but not intuit. AI in marketing should be used to analyze, predict, draw insights, respond, personalize, dialog, optimize, and then learn from all this and improve the next time. This is how AI will become an essential tool for marketers.