Chatbots are not responsive. Here’s how they work.

Microsoft released a new version of its Bing search engine last week, and unlike a regular search engine, it includes a chatbot that can answer questions in clear, concise prose.

Since then, people have noticed that some of what the Bing chatbot generates is inaccurate, misleading, and downright bizarre, raising concerns that it has become sentient or aware of the world around it.

This is not the case. And to understand why, it’s important to know how chatbots actually work.

Is the chatbot alive?

No. Let’s say it again: No!

In June, a Google engineer, Blake Lemoine, claimed that similar chatbot technology being tested at Google was susceptible. It’s wrong. Chatbots aren’t aware and aren’t intelligent – at least not in the way that humans are intelligent.

Why does he seem alive then?

Let’s take a step back. The Bing chatbot is powered by a kind of artificial intelligence called a neural network. It may look like a computerized brain, but the term is misleading.

A neural network is just a mathematical system that acquires skills by analyzing large amounts of numerical data. When a neural network examines thousands of photos of cats, for example, it can learn to recognize a cat.

Most people use neural networks on a daily basis. It is the technology that identifies people, pets and other objects in images posted to internet services such as Google Photos. It allows Siri and Alexa, the talking voice assistants from Apple and Amazon, to recognize the words you speak. And that’s what translates between English and Spanish on services like Google Translate.

Neural networks are very good at mimicking the way humans use language. And it can mislead us into thinking the technology is more powerful than it really is.

How exactly do neural networks mimic human language?

About five years ago, researchers from companies like Google and OpenAI, a San Francisco startup that recently launched the popular chatbot ChatGPT, began building neural networks that learned massive amounts of text. digital, including books, Wikipedia articles, chat logs, and all sorts of other things published on the Internet.

These neural networks are known as large language models. They are able to use these mounds of data to construct what might be called a mathematical map of human language. Using this map, neural networks can perform many different tasks, such as writing their own tweets, composing speeches, generating computer programs, and yes, having a conversation.

These large language models have proven useful. Microsoft offers a tool, Copilot, which is built on a large language model and can suggest the next line of code when computer programmers create software applications, much like auto-completion tools suggest the next word when you type texts or emails.

Other companies offer similar technology that can generate marketing materials, emails, and other texts. This type of technology is also known as generative AI.

Now companies are rolling out versions of this that you can argue with?

Exactly. In November, OpenAI released ChatGPT, the first time the general public got a taste of it. People were amazed – and rightly so.

These chatbots don’t chat exactly like a human, but they often seem to. They can also write essays, poetry, and riffs on almost any topic that comes their way.

Why are they wrong?

Because they learn on the Internet. Think about the amount of misinformation and other garbage on the web.

These systems do not repeat verbatim what is on the Internet. Based on what they have learned, they produce a new text themselves, in what artificial intelligence researchers call a “hallucination”.

This is why chatbots can give you different answers if you ask the same question twice. They will say anything, whether it is based on reality or not.

If chatbots “hallucinate”, doesn’t that make them sensitive?

Artificial intelligence researchers love to use terms that make these systems look human. But hallucinating is just a catchy term for “they make stuff up.”

It sounds scary and dangerous, but that doesn’t mean the technology is somehow alive or aware of its surroundings. It simply generates text using templates found on the internet. In many cases, it mixes and matches patterns in surprising and disturbing ways. But he is not aware of what he is doing. He can’t reason like humans.

Can’t companies stop chatbots from acting weird?

They try.

With ChatGPT, OpenAI tried to control the behavior of the technology. As a small group of people tested the system privately, OpenAI asked them to rate its responses. Were they helpful? Were they truthful? Then OpenAI used those ratings to fine-tune the system and define more precisely what it would and wouldn’t do.

But such techniques are not perfect. Scientists today don’t know how to build totally truthful systems. They can limit inaccuracies and quirks, but they can’t stop them. One of the ways to control strange behavior is to keep conversations short.

But chatbots will always spit things out that aren’t true. And as more companies start to deploy these kinds of bots, not everyone will be good at controlling what they can and can’t do.

The Bottom Line: Don’t believe everything a chatbot tells you.