The failed promise of Chatbots and how to fix it

I keep doing it, I keep clicking on that little icon in the bottom right corner.

I can't resist the temptation…

The temptation that just maybe, this time, fingers crossed, the chatbot could resolve my issue.

But I'm let down each time and I don't learn.

‍

Why do so many people still gravitate toward the chatbot whilst 99% of them suck?

‍

I think it's because of the promise of what could be.

The hope that the company I'm dealing with can infer my intent or my problem and resolve it immediately.

This promise was betrayed by the first wave of chatbot technology, and it's easy to see why.

‍

The majority of chatbots contain some combination of the following :

keyword matching, picking up a word in your sentence and then following a deterministic preprogrammed answer for that

asking a series of questions that equate to a conversational style menu, nothing more that a menu masquerading as a chatbot

incessantly suggesting "self-help" articles that I need to go read

capturing information 1 part at the time, even if I input "No you misunderstood me", resulting in my name being "No you misunderstood me" for the rest of the conversation

‍

We can summarize chatbots into a 2 categories :

a search tool, redirecting to self-help articles
a menu tool

‍

Hardly anywhere is there issue resolution.

But we're now in a new age, the age of LLMs.

‍

LLMs bring us into a position where we can finally fulfill this promise of chatbots.

Alas I think it might be too late to rescue the word "chatbot" from its sullied past, so at Stubber we call them "Chat Agents" or some variation of that.

‍

What do LLMs do that allow us to fulfill this promise?

‍

1. Inferring human intent

LLMs have the ability to infer the reason you are using the chat immediately, within 1 or 2 sentences.

Just tell the LLM what you want and it can narrow the range of options immediately.

This negates the need for : keyword matching, menu based programming, data parsing issues

2. Resolving the issue

The next thing that LLMs allow is actually reading the documentation and working through the issue with you.

This gets rid of the need to refer to some self-help article and ACTUALLY help you resolve the issue.

If we zoom up to the conceptual level.

What are humans doing when we're building menus, apps, portals?

We're trying to derive the possible "intents" of the users of apps & portals and then build an "intent navigation system".

With LLMs we can skip all this unnecessary work.

We can use the LLM to just get to the intent immediately.

‍

My view is that LLMs will usher in a new way of using computer interfaces, where we use our "default human communication mode" => natural language.

‍

This is the article intro text

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Duis curs justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristiq.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis curs justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Duis curs justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristiq.

September 14, 2024

Werner Stucky

The failed promise of Chatbots and how to fix it

Why do so many people still gravitate toward the chatbot whilst 99% of them suck?

What do LLMs do that allow us to fulfill this promise?

My view is that LLMs will usher in a new way of using computer interfaces, where we use our "default human communication mode" => natural language.

‍

This is the article intro text

Other Articles

Evolving Technology vs New Technology

The failed promise of Chatbots and how to fix it

Why we need a "thought" role in LLMs