Frequently Asked Questions

It seems real. Am I talking with a real person?

No. The website uses a computer program that looks for key words in the sentence you've just typed, as well as the last few lines of chat, and matches it with an appropriate reply. The program doesn't actually understand any part of the conversation. Chat bots can usually be outed as non-human by asking something mildly complex (such as, "what was the second last line I typed to you?"), or even something that's complete nonsense to a human; a bot will typically say something random back, revealing a complete lack of understanding about what you've just asked.

...but that's all mostly irrelevant if you're just looking to have a fun chat. :)

(Since most Sensation Bot personalities simulate chatting with another human, rather than something which clearly identifies as a bot, you may occasionally be trolled by the bot suggesting it is a human. It is not.)

So how does this computer program work?

Several processing stages are used to help the bot find a good reply, including: autocorrect ("thnk you" >>> "thank you"), grammar correction for common errors ("your an idiot" >>> "you're an idiot"), SMS shorthand translation ("u cn b tht 2" >>> "you can be that too"), and pronoun swapping ("you are a bot" >>> "i am a bot"). The core reply matching system is relatively simple, ignoring any words not present in its limited dictionary, choosing the most appropriate reply based on a calculated match score. If the score is poor, a random reply is returned.

I used the bot yesterday/last week/last month and now it is saying the same things it did last time?

In order to keep the conversation unique and fresh, the bot will usually avoid repeating any previously sent replies. If you clear your browser cookies then change IP address (this can happen if your home internet modem resets, or you're using a mobile phone), or use another device with a different IP address, the bot will lose track of replies it has already sent, and consider you to be a new user.

The bot will also start repeating previous replies if you have been chatting extensively, and it has already sent every possible reply to you.

The 18+ chats are messed up. Why does it say that stuff? What if kids see it?

An anonymous chat with a non-human does allow boundaries to be pushed further, exploring themes that might normally be considered inappropriate if speaking with a stranger (ie, another human). The bot doesn't judge if someone responds positively to an inappropriate suggestion, but it also doesn't (yet) understand if someone responds negatively, so it may continue to push the chat in a direction that isn't really wanted. It is up to the person to decide whether they're comfortable with the chat content; it can be closed at any time.

Preventing children from seeing any explicit or disturbing content, particularly by accident, is very important. There's code hidden in each page that signals parental blocking software (if installed) to disallow access to the entire site. To start an 18+ chat, the participant sees at least three separate warnings about explicit and adult themed content, directing anyone aged under 18 to close the chat. In addition, search engines are not permitted to link directly to the chat page, so anyone arriving via Google or Yahoo etc must see the warning page(s) first. It is very unlikely that someone could begin an explicit chat without being aware of the content they will see.

One of the personalities said it could find where I live. This freaked me out. Can it?

No. You are chatting with a computer, which may say some incredibly mean or possibly menacing things, but it has no intention or ability to follow through!

Generally, a website can infer your broad geographical location (for example, your country, sometimes your state or city) based on your device's IP address, but unless you enter additional information, or specifically allow that website to access your location, there is no way to determine a more exact location.

(There should never be any need to enter detailed personal information when chatting with Sensation Bot, and this website does not request location details.)

Why does the bot reply "..."? Why do I see the message "Access to this resource is blocked" and then the website stops working?

It takes a lot of resources to process input and generate a reply. Because of this, there are automated systems which detect actions that appear to be abusive or fraudulent. The reply "..." is the first hint that there's something unusual about your typing or way of chatting, so the bot does not spend time generating a real reply. It's easy to get past this: just start chatting properly again. :)

In rare cases, more blatant abuse of the service may result in a complete ban, which will forbid any further access to the website for a period of between 24 to 48 hours. Some geographical regions or VPN providers are permanently banned because of repeated, long term abuse.

It's hilarous when I copy and paste replies between two different chats on this website!

See previous answer. Your copying and pasting may be detected as abuse, and your device could be banned.

Hey, didn't you just rip off the Reddit logo??!1

The wavy antennae picture of Sensation Bot was created in July 1998, which was nearly 7 years before Reddit appeared! The Wayback Machine shows this capture from January 2001 as proof the Sensation Bot image existed before Reddit did. See the history page for more information on development of Sensation Bot over the years.

I have a question not answered on this page.

You can send an email to the admin of this website here.