On a LinkedIn post praising Woodland’s products, the company’s AI took it as a complaint and apologized. This automated response has highlighted the shortcomings and risks of AI-based customer service.
Nowadays is the era of generative AI, so companies are turning to automated replies on websites and social media to answer customer questions and feedback. Due to one such AI reply, Woodland Company, famous for shoes and clothes, is in trouble. Subin Abid, co-founder and chief product officer of a startup company named ZeroWatt Energy, shared a post on LinkedIn praising the durability of Woodland’s products, to which the AI gave a strange reply. This incident is a classic example of how no matter how good a company’s product is, how big things can go wrong if its online customer service is based on AI.
Subin Abid’s LinkedIn post was something like this
‘Dear Woodland, I don’t understand how you make money from customers. I purchased a belt from Woodland in 2013 and am still using it today. The slippers bought in 2021 still look like new. I think I could use them for a few more years. My previous Woodland shoes lasted over five years (though the new ones were a bit of a disappointment). When Woodland products are so sustainable, how long will it take for people to come back to you to buy a Woodland product?’ Expressing surprise, Subin Abid wrote in his LinkedIn post. The company’s automated response to this praise-filled post of Subin Abid was very strange. Woodland’s response to Subin Abid’s LinkedIn post began with the words, ‘Sorry for the inconvenience caused to you.’
Error in AI response
‘Hi Subin, we sincerely apologize for the inconvenience caused to you. Please send your concerns (along with copy/images of invoice) to [email protected] with your address. The company will respond within 24-48 hours’ – this was the reply from Woodland officials. But, many people in the comment box reminded Woodland that this was an automated reply and the answer was given without understanding the meaning of what Subin Abid had written. One user jokingly commented, ‘When such mistakes happen again and again, let’s speed up the hiring of AI by eliminating more humans from the customer service department.’ Another person wrote below the LinkedIn post, ‘The company doesn’t even have the sense that anyone is praising their products.’ Many people are considering this wrong answer as a reminder that even though AI is helpful, it can also make mistakes.