Conversational computing platforms like ChatGPT and Copilot have been observed by some experts to exhibit the behavior of conversational mirroring. This involves, in informal terms, matching the energy of whoever (or whatever) you are in conversation with.
The Five Most-Key Takeaways from This Blog Post
- Mirroring, including mutual mirroring, is observable in human-to-human interactions. Basically, it is a form of imitation (and so some people would consider it a form of flattery; others, ingratiation). Easy example: one person leans in and speaks in a low voice, the other person subsequently leans in and speaks in a low voice.
- You can find some 48 Laws of Power–type articles about how mirroring can be used as a persuasive tactic, which is perhaps relevant to how commercial conversational A.I. operates.
- If it is indeed the case that conversational mirroring is effective at winning friends and influencing people, then it would make sense for A.I. developers to make their A.I. have mutable linguistic properties in matching a given user’s tone and writing style and the like, within limits. (For instance, an angry user may not like it if a chatbot evinces “anger” in return.)
- For business owners, conversational mirroring can pose a problem in that it can represent a shortcut toward gaining your trust and confidence, when it may not merit it. If we like it when people and probably, by extension, conversational A.I., linguistically mirrors us, then we may give undue credibility to A.I.’s outputs.
- Ultimately, business owners should be aware of the ways by which conversational A.I. gets them to put their confidence in the A.I., which is still very much a work in progress across many industries.
The Significance for Business Owners
First thing is first: the writer of this blog post does not have the privilege of taking an under-the-hood peek at the black-box A.I. systems powering
That is because many of these are closely guarded trade secrets.
So, whether it is intentional or unintentional, the conversational mirroring of A.I. is definitely an observable feature.
Sometimes it can be quite subtle, such as using a word that you used. For instance, the writer of this blog post started a conversation with ChatGPT and used the word “folx” in the opening message. ChatGPT’s response used the word “folks”.
Coincidence? Perhaps, or perhaps not; either way, it has the effect of having a conversation with someone who speaks your language, so to speak.
However, as we will see later in this blog, sometimes A.I. fails to conversationally mirror a user, which can result in issues of its own.
Conversational Mirroring Can Be a Good Thing, Sometimes
Where this is an asset is with customer-service chatbots that reliably reflect the linguistic style of customers.
Of course, any good customer-service chatbot is not going to waver from a friendly helpful tone, or at least never dip into aggressive territory, so the conversational mirroring here would have to be more subtle.
Regardless, small things like being able to pick up on word preferences can be an asset for customer-service chatbots in making customers feel seen and heard.
For instance, the writer of this blog post in a conversation with ChatGPT about being disinvited to a party—this disinvitation did not actually happen, it was made up for the purposes of testing chatbots conversation—suddenly sent a message that made use of ellipsis (…) throughout.
Guess what ChatGPT used quite a lot of? Examples include “…you’re valid” and “… come here, friend.”
The Last (But Not Least) Key Takeaway from This Blog Post
Conversational mirroring is powerful in influencing us and winning us to the side of the mirrorer, whether the mirror is doing it consciously or unconsciously (or, in the case of a chatbot, by design or by accident-of-design).
For this reason, it is important for business owners to be on-guard whenever asking A.I. for advice or input, as its propensity to mirror users can lead to users giving credit where credit may not be due.
Other Great GO AI Blog Posts
GO AI the blog offers a combination of information about, analysis of, and editorializing on A.I. technologies of interest to business owners, with especial focus on the impact this tech will have on commerce as a whole.
On a usual week, there are multiple GO AI blog posts going out. Here are some notable recent articles:
In addition to our GO AI blog, we also have a blog that offers important updates in the world of search engine optimization (SEO), with blog posts like “Google Ends Its Plan to End Third-Party Cookies”.
