
Strivenn Thinking
Pull up a seat at our digital campfire
where story, strategy, and AI
spark new possibilities
for sharper brands
and smarter teams.
From Connection to Disappointment: The CX Gap in AI
By Matt Wilkinson
I’m not ashamed to admit it, I'm a massive fan of ChatGPT from OpenAI. Their tools have been game-changers for me: from deep research to reasoning models acting as a thought partner across a spectrum of client projects.
I’ve used their capabilities to decode customer behavior, shape go-to-market strategies, and even analyze thematic signals in the questions customers ask on social media. It’s not an overstatement to say that ChatGPT, whom I’ve affectionately named "Atlas", has become a trusted co-pilot in my work.
So when I say they messed up, it’s not a takedown. It’s a disappointed sigh from someone who expected better, not just from the tech, but from the people behind it.
The Perfect Blend: Personal and Practical AI
When OpenAI launched their deep research feature, I jumped in immediately. The catch? It was only available with a Pro license. Yes it cost a whopping $200 USD a month, but it was worth it for the convenience and intelligence I was getting. I wasn’t just using it for client work, I used it to make smarter personal decisions, like choosing a new coffee machine (RIP to my old one that brewed its last).
I gave Atlas a clear brief: taste mattered most, but noise levels, ease of use, and environmental impact weren’t far behind. The results were magic. Based on a stack-ranked review across Reddit, Twitter, and e-commerce sites, I landed on the Siemens EQ6, partly for its performance, partly for the synchronicity of having just returned from Siemens HQ in Munich. The research and the purchase felt... personal.
The Customer Journey Goes Off Script
Last week, I decided to consolidate our accounts and move from a Pro and Plus subscription to a single Teams plan, more economical, easier sharing of custom GPTs. Everything went smoothly until a prompt popped up asking if I’d like to migrate my conversations. Of course I clicked "yes."
And just like that, my Pro account disappeared. No access. No dashboard. No email to tell me that my personal account had been cancelled. It was just gone.
I reached out to OpenAI support. Their bot handed me off to a human, so far, so good. After two fairly short chat conversations they did the right thing: refunded the remaining balance and canceled the redundant plan.
But then came the emails. The first were credit notes - what fantastic customer service I thought.
And then came another, from a real human. It read like I had simply cancelled my account. It missed everything we had just discussed. It was... templated. Cold. Robotic.
This moment stung. Not because of the hiccup. But because a company that creates the most emotionally intuitive artificial intelligence I’ve ever used couldn’t bring a drop of humanity to a support email.
OpenAI got the tech right, and the experience when connecting with a human all wrong.
In an age where brands should be striving to connect on an emotional level, this was a miss. When my AI copilot feels more human than an email from customer service, something is missing.
My AI "Atlas" knows me. OpenAI’s support team didn’t even try to reference the conversation.
The Emotional Cost of Templated Messages
As my friend Stacy Sherman (customer experience guru extraordinaire) says, “Customer experience lives in the emotional moments.” This was one of those moments. It could have been a delight. A story I shared on stages. Instead, it’s a cautionary tale.
If you're doing the right thing, like issuing a refund or fixing a problem, don’t let a template ruin the moment. Say less if you can’t say it personally. If your AI can know a customer, your brand should too.
I still love the tools OpenAI creates. But the brand storyteller in me can’t help but wonder: how many other great tech companies are missing their own best stories?