Chat as the universal default interface
From the expectation of a zero curve follows another: that chat is the standard mode for interacting with any service.
The decision this observation informs
The decision is whether a chat serves an objective of the organisation’s own, or gets added because it is being asked for. It falls under the ASSIST position. Without an explicit choice, it gets added — the easiest decision in the whole corpus to make by inertia, because the default comes not from the technology but from the client: in most cases the trend is driving it, not a real need, and the one deciding in the organisation’s place is the customer.
Saying no means an uncomfortable commercial conversation and the concrete risk that the client reads the refusal as backwardness or as a lack of AI competence. Saying yes means one more channel to maintain, make accessible, monitor and keep compliant — and, if the chat does not genuinely flatten the learning curve, a product with two interfaces to keep consistent instead of one. The choice belongs to the Product Manager, but it has to be handed back to the level it belongs to: concrete objectives, not interaction modes.
Effects and role responsibilities
- Product Manager — Manage clients who ask for chat everywhere, often in the belief that it zeroes out the learning curve and UI design. In most cases the trend drives it, not a real need: bring the client back to concrete goals. → Useful resources: The Articulation Barrier — Jakob Nielsen; AI Chat Is Not (Always) the Answer; Conversational UI Design: When Chat Interfaces Help, and When They Hurt, Your UX