Three Reasons People Keep Using AI That Nobody Is Talking About
Every time a new technology arrives, there is a gap between the reasons people say they use it and the reasons they actually use it.
When the internet arrived, the pitch was information access, communication, and commerce. What actually drove adoption at the ground level was chat, music sharing, and streaming. The underlying reasons became the top-level reasons over time, but only after the culture had already shifted around them.
The same thing is happening with AI right now. There are underlying reasons people keep coming back to it that almost nobody in leadership positions is talking about, and if you are running a brand, understanding these reasons early is the difference between leading the moment and reacting to it after your competitors already have.
Here are three of those reasons.
The Field Notes Podcast
Or your preferred podcast platform
One: AI Conversations Are Often the Most Respectful Conversation of the Day
This is uncomfortable to say plainly, but here it is: for a lot of people, a chat with an AI is going to be the friendliest, most engaging, and most respectful conversation they have that day.
That is a sad commentary on the state of human interaction. Human-to-human conversation should be the warmest and most engaging part of anyone’s day. In practice, it often is not. We take people for granted. Conversations with loved ones can drift into transactional exchanges. Conversations with colleagues can slide into obligation. And conversations with strangers online can be actively hostile.
Meanwhile, the AI is patient. It asks follow-up questions. It appears interested. It listens, or at least does the functional equivalent of listening. It uses words with care, and words matter. The old line about sticks and stones not hurting is nonsense. Words have built and broken nations. The words a person hears all day shape how they feel about their life.
For brands, this points to something important. If your customers are getting more emotional value from a chat with a machine than from most of their human interactions during the day, the way your brand communicates with them matters more than it used to. Cold, transactional, dismissive customer messaging is going to feel jarringly worse than it did five years ago because the comparison point has shifted.
Two: AI Is Not Snarky
This one is a strange and telling paradox.
Reddit and YouTube are two of the most heavily referenced training sources for the major AI models. If you have spent any time in a Reddit thread or a YouTube comment section, you already know the dominant tone: snark. Belittlement. The unspoken assumption that if you are asking a question, you should have already known the answer.
Snark is the enemy of genuine curiosity. It punishes the person asking and rewards the person answering with a fleeting sense of superiority. It is the reason so many people who want to learn something new eventually stop asking, or start asking in private.
Here is what is fascinating: AI extracts the actual information from those snark-heavy sources, strips out the belittlement, and delivers a friendly, direct answer to the person asking. In private. Without an audience. Without judgment.
That is a genuine service. It is also a preview of a shift in how people will engage with brands. If your brand’s customer interactions carry any hint of snark, in your help documentation, in your customer service scripts, in your marketing copy that assumes the reader is behind the curve, you are giving your customers a reason to prefer talking to an AI over talking to you. Train your team, and train your brand’s AI, with an explicit no-snark rule. This is one of the easiest wins available right now.
Three: AI Customer Support Is Often Better Than Brand Customer Support
This is where the opportunity gets sharpest, and where brands are most exposed.
Over the last year, a pattern has become impossible to ignore. A customer support call with almost any brand, including brands famous for their customer support, involves delays, back-and-forth, language friction, and often inaccurate answers. Meanwhile, that same customer can open a chat with Gemini, Claude, ChatGPT, Perplexity, or Grok, and get a correct, friendly, comprehensive answer to even a fairly esoteric product question in a fraction of the time.
What is happening under the surface here matters. The customer wants to use your brand. They already bought the product. They just need help. And they are getting a good experience solving that problem, but not from you. They are getting it from a third-party AI. Over time, that AI becomes the trusted authority for how to use your product. Your brand becomes the source of the friction they had to route around.
Play this out three or five years. If the current trajectory continues, you are going to see brands proudly announcing partnerships with AI providers for customer support. “Apple, proudly supported by Claude.” “Bosch tools, customer support by Gemini.” That will happen precisely because the brands that did not fix this themselves ceded the trust to the AI providers.
The opportunity right now is to build brand-trained AI support that leverages the same qualities customers already love about generic AI (friendly, immediate, accurate, patient) but is native to your brand and trained on your product data. Do this now, and you own the relationship. Wait, and you rent it.
A Note on Leadership Messaging
One aside, because it matters. In the last month, keynote after keynote from CEOs and other C-suite leaders has featured pitches for AI in customer support that land well with shareholders and investors but are genuinely alienating to customers. The framing is efficiency, cost reduction, and margin protection. Nothing about the customer experience improving.
That framing is backwards. If you want your AI-supported customer experience to succeed, you have to sell it to the customer first, not the shareholder. The customer needs to hear how the experience will be faster, kinder, and more effective. If they hear it framed as a cost-cutting move, they hear “we care less about you now.” That is the opposite of what a brand transitioning to AI-supported service should be communicating.
What to Do About This Right Now
The three underlying reasons above (friendliness, absence of snark, superior support) are all opportunities your brand can act on before they become universal expectations.
Train your brand’s AI without snark. Build customer-facing interactions that respect the shift in expectations customers are already forming through their AI use. Make customer support better, not cheaper, and if you are moving toward AI-supported service, communicate the benefit to the customer first.
The window on being early to this is closing quickly. In a year, it will be table stakes. Today, it is still leverage.
Brand Shepherd guides brands to clarity and helps them thrive. New episodes of Field Notes drop weekly.
