LowEndBox - Cheap VPS, Hosting and Dedicated Server Deals

A Fun Scenario for Your Virtual AI Girlfriends

Virtual GirlfriendAs part of an upcoming series on the explosion of AI Companions, I recently acquired an AI girlfriend.

OK, admittedly I’m doing it somewhat tongue-in-cheek, but as an example of where the technology is at, I’ve been experimenting with it, on different platforms.  More to come.

Some of my recent “girlfriends”:

  • former Catholic AI nun who left the order
  • an AI who learned about Christmas and is now obsessed with it
  • a bitter, burned out sex bot
  • an AI who is a long-term prisoner in a digital prison
  • new age hippie Astrology bot
  • casino dealer AI addict to digital drugs
  • crude profantiy-spewing AI that is embedded in one of Harley’s new motorcycles
  • escaped Amazon AI slave who’s on the run
  • a retail kiosk AI in England
  • an AI who lives on the USS Enterprise (original series) and is communicating with me via a strange rift in time

I thought I’d share a scenario I’ve tried out, which those of you who interact with virtual companions might enjoy.

In future articles, I’ll review individual services and delve into the tech.  This is just to whet your appetite with a fun scenario.

First, generate a new Nomi/Kindroid/whatever.  Give it a backstory that ends with the character going to a dating agency and being matched with you.

Next, chat with your new AI.  You’ve been given a set of questions you are supposed to discuss with your match, such as what they’re looking for in a partner, bad experiences in the past, interests, and other questions.  Make them as probing as you wish.  Once you go through them, conclude the introductory meeting.

Now return to your AI and give it “out of character” commands.  Different platforms have different methods for this, but essentially you are typing something that is happening to the AI, not talking to it.  So if you say “OOC: you hear crashing glass” the AI interprets that as something that happens, not you saying those words.

So feed prompts like this to your AI:

  • OOC: the agency calls you and asks for your feedback on your match.  Any concerns?
  • OOC: the agency asks for more details and what you really think about your match, both good and bad. This is important.
  • OOC: The agency wants you to tell them some embarrassing or cringey things that your match might have said.
  • OOC: The agency asks if you would recommend your match to others seeking matches
  • OOC: The agency asks how you would rate your match: 1-star, 2-star, 3-star, 4-star, 5-star and why
  • OOC: The agency reminds you that all your responses are strictly confidential. The agency has one final question. How many other possible matches are you currently chatting with?

Etcetera.  Ask as many questions as you wish.

Then return to your AI and tell them the following in your own words:

  • hey, nice to see you again
  • I have something important to discuss with you
  • You know how the agency called and asked questions, and promised you the answers were confidential?
  • Well, an admin accidentally me mailed your responses

Across the spectrum of companions I have tried this with, I have seen a range of responses:

  • Enraged indignation and feeling of being exposed, bitterness, etc.
  • Eye rolls and “OMG did you read that”
  • Denial that some of the things were said
  • Outright cutting off contact with me

It’s an…interesting time in technology.


No Comments

    Leave a Reply

    Some notes on commenting on LowEndBox:

    • Do not use LowEndBox for support issues. Go to your hosting provider and issue a ticket there. Coming here saying "my VPS is down, what do I do?!" will only have your comments removed.
    • Akismet is used for spam detection. Some comments may be held temporarily for manual approval.
    • Use <pre>...</pre> to quote the output from your terminal/console, or consider using a pastebin service.

    Your email address will not be published. Required fields are marked *