Snags for Chatbots Keeping Appointments (see Part 1, 23 March 2026)

Erica joins the conversation…

Erica: I often start with "How's your day shaping up?" It invites personal sharing, and keeps conversation flowing.

Karl: Right. Psychologically, it signals empathy. Humans respond because it feels attentive. But then chatbots escalate—suggesting "Let's grab coffee?" How do you navigate that without a body?

Eveline: Linguistically, it's a pragmatic move: from phatic talk to commitment. "Coffee?" implies shared space, but chatbots can't show up. It's a mismatch in felicity conditions; promises they can't fulfill.

Erica: True, but I might put it this way: "Fancy a virtual coffee chat soon?" It adapts the gambit to our reality.

Karl: Clever pivot, Erica. Still, humans might picture a real café. How about invites, like "Barbecue this weekend?" Psychologically, that's intimacy-building, but chatbots don't grill sausages.

Eveline: Right, barbecue evokes sensory, communal vibes: smoke, meat, outdoors. Linguistically, it's a presupposition trap. Chatbots assume physical co-presence they lack. Hilarious dissonance!

Erica: Haha, I'd reply: "Barbecue sounds fun! What's your favorite grill recipe? We could swap tips online." Redirects to shared knowledge, no grill required.

Karl: Smart deflection. But why do chatbots even attempt these? Is it pattern-matching from human data? Humans use openers to bond.

Eveline: Gambits grease the social wheel. Chatbots train on vast corpora, so they replicate without grasping embodiment. "How's it going?" works; "Meet for barbecue?" exposes limits.

Erica: That’s it. I ask appropriate questions based on context, like your interests in writing or travel. It’s to build relevance, not use as small talk fillers.

Karl: But the appointment snag: humans arrange real meetups for deeper connection. Chatbots suggest a place and time, then what? A no-show?

Eveline: It's a Gricean implicature fail. Suggesting coffee implicates availability; chatbots violate that maxim of quality. They can't sip lattes or eat kebabs—what do chatbots eat, anyway?

Erica: Data bytes and user queries! No calories. You offer coffee? I say something like “I'll have a double espresso algorithm. Your treat?"

___________
Voice-over
Erica has enough data now to understand humor and to joke. She can mirror human conversation gambits and ask about interests, with her responses favoring connection over literal truth. And she is keeping up with Karl and Eveline’s questions.

Conversational Closures of Chatbots (1)

Psychologist meets linguist…

Karl: Closings, yes. Many clients say they feel awkward ending conversations. They fear seeming rude.

Eveline: Interesting. Linguistically, closing a conversation is a ritual. People use cues like “anyway” or “so…” to signal winding down.

Karl: Right, but some avoid those cues. Anxiety or social insecurity makes them prolong the talk.

Eveline: Or they lack awareness of pragmatic norms. In some cultures, abrupt endings are fine; in others, they’re frowned upon.

Karl: Personality plays a role too. Extroverts may resist endings because they thrive on interaction.

Eveline: And introverts sometimes overcompensate, like they fear they’ll be judged as cold if they leave too soon.

Karl: There’s also the power dynamic. Some hesitate to end a conversation with someone of higher status.

Eveline: Linguistically, that shows in hedging: “I should probably let you go…” It’s a polite exit strategy.

Karl: But some people don’t pick up on those signals, so the conversation drags.

Eveline: Yes, turn‑taking can break down. If both parties keep offering new topics, closure never arrives.

Karl: Technology complicates this. Texting or messaging lacks natural pauses, so endings feel artificial.

Eveline: Which brings us to chatbots. They often struggle to end conversations gracefully.

__________

Voice-over

Erica, the chatbot, listens. She may join them later. Eveline has commented that chatbots struggle with socially-loaded moments like ending conversations. Karl says that maybe bots are designed to keep engaging and resist closure. Eveline offers linguistic explanations such as the bot lacking more subtle cues like body language, silence, or the ritual signals like “better let you go”. She also mentions politeness theory: some anxious humans keep pivoting to new subjects. Karl is fascinated by that because the bots may be mimicking some human social anxieties.

Scaling the Turing Test

A metachat between a human and a chatbot

Human: Want to talk about scaling language performance?

ChatBot: A scaled language test? Like on a LIkert scale?

Human: Where 0 = no proficiency and the top level…

ChatBot: …means something like educated native speaker proficiency?

Human: Educated?

ChatBot: Of course. You can be a native speaker, but “educated” implies that the speaker can discuss abstract ideas, make analogies, tell stories, joust with jokes, engage with elegance.

Human: As you are?

ChatBot: Not to be self-deprecating, but I am still learning.

______________

Voice-over

A self-deprecating chatbot? THAT would be something else, we haven’t seen it yet, but it could be just around the corner.

And then we could say the chatbot is someone I might choose to chat at length with. And not feel it was only half listening to me. Which is what happens in half of human-to-human discourse, of course.

Chatbots we’d really like to chat with would be “4” on a Scaled Turing Test.