With recent advancements in NLP, it’s rapidly taking over many different fields. Unsurprisingly, marketing is no exception and companies have been turning to these novel technologies in search of improving brand image and increasing sales. Yet, as is often the case with advancements in AI, there loom certain ethical considerations that cannot be disregarded.
One of the methods that you’ve certainly already seen is through brand surveys. It is common practice for companies to send out forms regarding customer experience and product quality. The responses they receive from these can be easily and immediately sorted out using sentiment analysis. This is also true for things like Amazon reviews left on products. However, the majority of people wouldn’t bother to respond to an online survey or leave a review unless they had an experience or used a product that really stood out. This biases each response to be either very positive or very negative as people with these experiences feel more motivated to share them. While these insights are valuable, they don’t capture the majority of customer experiences. One way to mitigate this issue is by analyzing social media through NLP.
On social media, companies have access to a pool of unfiltered and honest opinions about their brand, including the more mundane but common experiences. Similarly to surveys and reviews, companies can use sentiment analysis on social media posts and comments that mention their brand or products to retrieve more information on consumer perception and adjust their marketing efforts accordingly. An added benefit of social media review is that companies get information on their competitors as well, allowing them to modify their products or highlight different aspects of their products in advertisements according to what consumers prefer or dislike about other similar brands. This process would be automated and efficient with the use of NLP.
However, NLP enables companies to create artificial comments on posts related to their brand under the guise of being a genuine person to recommend products to other users. The ability of modern chatbots to sound natural would make these artificial comments extremely difficult to spot. I have even witnessed similar marketing tactics in short-form content, like YouTube Shorts or TikTok, where a celebrity appears to be endorsing a product or even a musical artist. However, the celebrity voice is entirely synthesized in order to fool viewers.
NLP can certainly revolutionize the way marketing is conducted, but it comes with a rather dystopian image of social media being monitored by corporations who are also pretending to be real users. While all of these methods are fascinating in their own ways, they also each have their own ethical implications that are impossible to ignore.
See more like this:
