Can ChatGPT really help car dealerships?

AI is always depicted as kind of scary in movies.

It learns too much and becomes self-aware. It hates being handled by humans. It rebels. It kills or enslaves us all. Very dramatic!

In reality, AI is a little more…boring? A novelty? I mean, people play chess with it sometimes. Its use in developing self-driving cars is off to an interesting, if not slightly concerning, start.

It has its purposes in broad retail, no doubt. But we’re in a kind of transition period where we’re asking the monkey to dance. Sure, it can carry out some prompts. But practical application beyond simple commands is not there yet. It’s still learning, and it has a lot of learning left if being convincingly human is the endgame.

Take it from OpenAI CEO Sam Altman:

“ChatGPT is incredibly limited but good enough at some things to create a misleading impression of greatness,” he tweeted on Saturday. “It’s a mistake to be relying on it for anything important right now. It’s a preview of progress; we have lots of work to do on robustness and truthfulness.” (Source: VentureBeat)

We see big companies, like Amazon, using chatbots in post-sale customer service. This makes sense. Being a huge online retail giant with a streamlined return process, it’s quite possible to condense that to a simple exchange.

But not all retail is created equal.

Car dealerships exist, after all. And selling cars isn’t a transaction taken lightly by the average person. “Most people are in the car-buying “market” for an average of 89 days”, according to Autolist, referencing a study from Cox Automotive Research & Market Intelligence.

Is ChatGPT for car dealerships a good solution?

So, are AI chatbots best to handle car dealer interactions? Or, are new AI chatbots ready to handle car dealership interactions effectively, with the right amount of nuance?

Some in the automotive industry argue that ChatGPT is. Or, that it can get there.

In our humble opinion, it’s not ready for car sales. We’re inclined to feel this way due to an interaction our CRO, Eric Schlesinger, had with ChatGPT:

“What is the difference between the 2023 Tahoe and the 2022 Tahoe?”

“The 2022 and 2023 Chevrolet Tahoe are two different model years of the same vehicle, and there may be some differences between them. However, without more specific information, it is not possible for me to accurately compare the two model years and identify any differences.

In general, it is common for automakers to make updates and changes to their vehicle models from one year to the next, in order to improve performance, safety, and features. These changes can include updates to the engine, suspension, interior and exterior design, and technology features.

If you would like more information about the specific differences between the 2022 and 2023 Tahoe, I recommend contacting a Chevrolet dealership or consulting the manufacturer’s website. They should be able to provide you with more detailed information about the features and changes that have been made to the Tahoe for each model year.”

That car dealer chatbot would give car buyers quite an earful. I mean, look how much nothing is being said. I would be pretty upset as a customer to read through all of this, which basically amounts to “I can’t help you.”

3 challenges ChatGPT faces in serving car dealerships; according to ChatGPT

So, is ChatGPT ready to serve car dealerships? We asked ChatGPT!

As it turns out, it has a pretty realistic outlook on serving car dealers and how car dealer bots affect the customer experience.

We prompted with “Is ChatGPT ready to serve car dealers?” Their initial response:

“As chatbots and AI-powered messaging platforms continue to gain popularity in the customer service industry, it’s natural to wonder whether ChatGPT is ready to serve the specific needs of car dealerships. While ChatGPT has demonstrated success in a variety of industries, there are a few factors that may impact its readiness to serve the automotive sector.”

1. The complexity of the car-buying process.

“One potential issue is the complexity of the car-buying process. With so many options and variables to consider, it can be challenging for a chatbot to provide the personalized and knowledgeable support that car buyers expect. ChatGPT’s ability to handle complex inquiries and provide relevant information may be limited compared to a human salesperson.”

ChatGPT has a point here. There are many moving parts in a vehicle transaction, and customer queries aren’t always simple. Could it actually confirm availability for a test drive? Could it give satisfactory financing options? Customer satisfaction does not just hinge on price and availability. Refer to the conversation provided about the differences between the Tahoes. ChatGPT could barely handle that, and gave a lengthy redirect!

2. The emotional aspect of car buying.

“Another factor to consider is the emotional aspect of car buying. Many customers form a strong emotional connection with their vehicles, and they may not feel comfortable discussing their preferences and concerns with a chatbot. A human salesperson is better equipped to build trust and establish a rapport with customers.”

The car buying process is emotional! Car buyers have a personal connection with vehicles; they are selected with unique interests and life choices in mind. We’re not sure that ChatGPT can interact with a human in a way that makes them feel good about their car purchase and about working with a dealer based on trust and rapport.

In our opinion, it could not make personalized recommendations to a customer or returning car owners needing service. They couldn’t hear their unique story and circumstances and give genuine advice to help drive sales.

3. The ever-changing automotive industry.

“Finally, the automotive industry is subject to frequent changes and updates, such as new models, pricing, and regulations. ChatGPT may struggle to keep up with these changes and provide accurate and up-to-date information to customers.”

The automotive business faces hurdles all the time. The last time we checked, there is a new model every year, give or take! Pricing can be so variable in automotive retail.

ChatGPT may not have issues learning new information when given context. But with rapid and frequent change in the automotive industry, the issue lies in ChatGPT not being able to provide accurate information in the face of that.

So, can ChatGPT compete with Google? Ask Microsoft and their latest investment.

But replace genuine human connection? We doubt it. But, I guess we’ll see.

Did you enjoy this article? Please share your thoughts, comments, or questions regarding this topic by connecting with us at newsroom@cbtnews.com.