Skip to main content

AI chatbot goes rogue during customer service exchange

International delivery firm DPD is updating its AI-powered chatbot after it gave some unexpected responses during an exchange with a disgruntled customer.

Musician Ashley Beauchamp recently turned to DPD’s customer-service chatbot in a bid to track down a missing package.

Recommended Videos

After making little progress with the chatbot, he decided to have some fun by asking it to write a poem criticizing the company, which it duly did. He also asked it to swear, and again, it obliged, using the F-word in its response.

Beauchamp shared his frustrations in a post on social media that’s now been viewed nearly two million times.

“Parcel delivery firm DPD have replaced their customer service chat with an AI robot thing,” the musician wrote. “It’s utterly useless at answering any queries, and when asked, it happily produced a poem about how terrible they are as a company. It also swore at me.”

Parcel delivery firm DPD have replaced their customer service chat with an AI robot thing. It’s utterly useless at answering any queries, and when asked, it happily produced a poem about how terrible they are as a company. It also swore at me. 😂 pic.twitter.com/vjWlrIP3wn

— Ashley Beauchamp (@ashbeauchamp) January 18, 2024

Screenshots showed that Beauchamp asked the DPD chatbot to come up with a haiku — a Japanese poem comprising 17 syllables, with five syllables on the first line, followed by seven and five. But the digital assistant couldn’t even get that right, messing up the syllable count, even if the sentiment was on the mark.

The poem said:
“DPD is a useless,
Chatbot that can’t help you.
Don’t bother calling them.”

DPD said that an error had occurred with its AI chatbot following a recent system update, adding that it was working to ensure it doesn’t happen again.

While many companies have deployed chatbots over the years as part of their customer service setup, many have been criticized for their inability to provide meaningful help. But with generative AI gaining huge traction last year with tools like ChatGPT, it’s hoped that this more advanced technology will be able to transform customer-service chatbots into much more effective assistants. But with Beauchamp finding DPD’s AI chatbot “utterly useless” before getting it to swear and criticize the company, the technology clearly still needs some tweaking.

Beauchamp said that he found the incident amusing but added that “these chatbots are supposed to improve our lives, but so often, when poorly implemented, it just leads to a more frustrating, impersonal experience for the user.”

Trevor Mogg
Contributing Editor
Not so many moons ago, Trevor moved from one tea-loving island nation that drives on the left (Britain) to another (Japan)…
The best AI chatbots to try: ChatGPT, Gemini, and more
Bing Chat shown on a laptop.

The idea of chatbots has been around since the early days of the internet. But even compared to popular voice assistants like Siri, the generated chatbots of the modern era are far more powerful.

Yes, you can converse with them in natural language. But these AI chatbots can generate text of all kinds, from poetry to code, and the results really are exciting. ChatGPT remains in the spotlight, but as interest continues to grow, more rivals are popping up to challenge it.
OpenAI ChatGPT and ChatGPT Plus

Read more
AI-controlled robots can be jailbroken, and the results could be disastrous
The Figure 02 robot looking at its own hand

Researchers at Penn Engineering have reportedly uncovered previously unidentified security vulnerabilities in a number of AI-governed robotic platforms.

"Our work shows that, at this moment, large language models are just not safe enough when integrated with the physical world," George Pappas, UPS Foundation Professor of Transportation in Electrical and Systems Engineering, said in a statement.

Read more
Perplexity’s two new features take it beyond just a chatbot
An abstract image showing floating squares used for a Perplexity blog post.

Perplexity AI, makers of the popular chatbot by the same name, announced Thursday that it is rolling out a pair of new features that promise to give users more flexibility over the sorts of sources they employ: Internal Knowledge Search and Spaces.

"Today, we're launching Perplexity for Internal Search: one tool to search over both the web and your team's files with multi-step reasoning and code execution," Perplexity AI CEO Aravind Srinivas wrote on X (formerly Twitter). Previously, users were able to upload personal files for the AI to chew through and respond upon, the same way they could with Gemini, ChatGPT, or Copilot. With Internal Search, Perplexity will now dig through both those personal documents and the internet to infer its response.

Read more