The Customer Still Gets a Vote
Plus: AI may reduce admin work. It creates a CX problem the second customers feel cornered.
Your daily signal on AI and CX — minus the hype.
📌 DCX Stat of the day: 36% of entrepreneurs say they would avoid relying on AI for decisions affecting customers, regardless of convenience. LegalZoom
In this issue:
→ Consent becomes the AI trust test
→ Customers still want judgment at risk
→ Readiness moves before go-live
→ Payment failures get flagged earlier
→ Produce quality gets machine vision
🔎 Deep dive
Consent is no longer a policy form
A Melbourne psychiatrist reportedly refused new patients unless they agreed to AI note-taking during sessions. The registration form told patients who objected to arrange for their doctor to refer them somewhere else. The Guardian also reported that two in five Australian GPs are now using AI scribes, with uptake doubling over the past 12 months.
That is where the customer feels the cost.
In mental health, the patient is not making a normal software preference choice. They are trying to get care. If declining AI means restarting a referral, waiting again, or walking away, the choice starts to feel like a toll booth. The business may see efficiency. The customer may feel pressure, exposure, and loss of control.
This will first appear in intake, consent language, privacy disclosures, provider scripts, and escalation paths. The executive implication is blunt: any AI tool that records, summarizes, routes, scores, or stores customer information is now part of the experience. Treating it like an internal productivity tool is how trust quietly leaks out of the journey.
📬 Copy-Paste Take
Before we roll AI into any customer conversation, we need to define the customer’s real choice: what they see, what they can decline, and what happens next. If declining creates a penalty, don’t call it consent.
OPERATOR PLAYBOOK
Pressure-test the opt-out
The issue is not whether the AI works. That only gets you to the starting line.
Audit every AI-assisted consent or disclosure flow for four things:
Where the customer first learns AI is involved.
Whether the choice is plain, visible, and timed correctly.
What happens if the customer says no.
Who owns recovery when the customer feels pushed.
Then test whether a reasonable customer could understand the choice without needing legal patience or CX-level stubbornness.
Ask your team: Where are we asking customers to accept AI because it makes our process easier?
Signal: The opt-out experience tells you whether the consent is real.
📈 Market Reality Check
Adoption is moving faster than judgment
LegalZoom’s survey of 1,000 aspiring founders and business owners gives this issue a wider frame. Among AI users, 42% use AI daily, 36% use it weekly, and 77% use it at least weekly.
That is the easy part to celebrate.
The useful part is where people draw the line. Respondents said they would avoid AI for high-risk legal or financial decisions, customer-impacting decisions, employee-impacting decisions, and situations involving regulatory penalties. People are not rejecting AI. They are refusing to hand over judgment when the consequence lands on a person.
That should shape how CX teams govern AI rollouts.
Convenience without judgment = risk transferred to the customer.
🧰 Tool Worth Knowing
Smart Role
What it does: Smart Role is a pre-production readiness layer for customer support teams. It helps companies build role-play scenarios, virtual customers, scorecards, and readiness gates so agents can practice before they touch live customers.
CX use case: Use it before new hire go-live, BPO transitions, seasonal staffing spikes, product launches, policy changes, and regulated support flows where “completed training” is weak evidence.
Worth watching because: Smart Role is solving the right problem: the gap between training completion and conversation readiness. Its readiness model is simple and useful: pull from real tickets, past QA misses, and SOPs, turn those into simulations, then certify readiness before go-live. Stop using customers as the final QA filter.
Bottom line: The scorecard matters. If the scenarios reflect real customer pressure, policy complexity, and recovery moments, this can prevent damage before it reaches the floor. If the scenarios are soft, you’ve automated a nicer version of the same hopeful sign-off.
NEW: The DCX AI Today - AI Tool Directory - If you lead a CX team and want a curated shortlist of tools worth evaluating, this is your starting point.
⚡ 90-Second CX Radar
Mastercard wants to catch payment failure before the customer feels it
Mastercard’s A2A Payment Intelligence predicts whether an account-to-account payment is likely to clear before it is initiated. Failed payments rarely feel like payment operations to customers. They feel like fees, service interruption, confusion, and one more reason to call.
Albertsons is putting AI upstream of the produce aisle
Albertsons launched an AI-powered Intelligent Quality Control tool for produce inspection. Customers do not care that the quality miss happened in the supply chain. They care that the berries went bad before the receipt stopped feeling new.
SmartBear flags the AI quality debt problem
SmartBear says 70% of software experts are concerned application quality is already suffering as AI speeds up development, and 60% have experienced quality issues because development moved faster than testing could validate. Faster shipping is not a CX win when the API, workflow, or edge case breaks in front of the customer.
🧭 Your Move
Pick one AI-assisted moment this week and look at it from the customer’s side of the glass.
Can they see AI is involved? Can they decline it? Can a frontline employee explain what happens next without sounding like they’re reading from a policy bunker? And if the customer says no, does the journey still work?
That last question is the one to sit with.
The customer does not experience your AI roadmap. They experience the choices your AI roadmap leaves them.
Until tomorrow,
Mark
👥 Share This Issue
Think of one person who’s wrestling with AI in CX right now
and forward this to them.
I’m obsessed with Wispr Flow Pro! Get a Free Month on me.
If someone forwarded this to you, they thought you needed to see it before your next AI planning meeting. Get your own copy.