• by Todd Wasserman , Staff Writer, October 3, 2022

If you’ve ever used a chatbot to reach a brand and wondered if you’re talking to a person or merely software, you’re not alone. Some 80% of people have interacted with a chatbot at some point, according to Userlike. Chatbots are able to handle full conversations about 70% of the time, per Comm100.

For many consumers, there often isn’t much of a difference anyway. Some 40% of consumers don’t care if there’s a human agent on the other end or not, which makes sense if you merely want an answer to a common question, according to HubSpot.

But there’s a dark side to chatbots. Roughly half of consumers polled by NewVoiceMedia believe that chatbots are used to prevent people from speaking to agents. Anyone who has screamed “Talk to a representative!” to an automated representative on the other end of the phone has experienced this sensation.

The pandemic accelerated the rise of chatbots. In a 2021 survey by Drift, 58% of respondents said they adopted a conversational marketing solution because of COVID.

Of course, AI is the backbone of most chatbot solutions, and AI is improving. Most recently, Alphabet-owned AI lab DeepMind unveiled Sparrow, an AI chatbot that uses a live Google search to give human-sounding answers to customers’ questions. The model managed to give plausible answers to factual questions—using information retrieved by the Internet—78% of the time, according to the company.

That remaining 22% is worrying to DeepMind researchers, according to Sara Hooker, who leads Cohere for AI, a nonprofit AI research lab. “For areas where human harm can be high if an agent answers, such as providing medical and financial advice, this may still feel to many like an unacceptably high failure rate,” Hooker told MIT Technology Review. “But the improvements are convincing and show clear benefits to human-guided optimization of dialogue agents in a large-language-model setting.”

Still, recent events show consumers should at least take chatbots’ comments with a grain of salt. This summer, Meta’s new chatbot, BlenderBot 3, made anti-Semitic remarks, according to CNN Business, and openly blasted Facebook, per Vice. At this current moment, marketers should greatly modify how chatbots interact with consumers and use chatbots with extreme caution.