Resurrecting the AI Expert System

Resurrecting the AI Expert System

The AI Expert System was one of our primary components back when we started AINIRO in addition to our AI chatbot technology. However, it was immature, it didn't really have many use cases, so to consolidate our portfolio and focus on what we were good at, we ditched it about 9 months ago.

With our newly added AI functions, this fundamentally changes, since I can imagine "a bajillion" use cases for creating password protected AI systems now. Below are some examples of things I can imagine that requires a private AI chatbot not accessible to the general public.

  • Email blaster that sends personalised marketing emails.
  • AI assistants that has access to things you don't want to publicly expose, such as your contacts, sending emails on your behalf, or your CRM system.
  • HR systems that matches resumes with job descriptions to filter out unqualified candidates.
  • CRM systems based upon natural language and AI instead of a UI.
  • Content production AI-based systems, integrated with your CMS, allowing you to automatically publish content.
  • Product analysis, allowing you to summarize products before purchasing to help you out during the purchasing process of some component.
  • Analysing documents and files, such as for instance PDF files, websites, etc.
  • Meeting preparation by scraping URLs of people you're about to have a meeting with, to help you prepare for the meeting.
  • Privately accessible cognitive assistance to your existing (humanly manned) customer service department.
  • Basically, anything you could possibly ever need AI to help you with - And it's all no-code and low-code due to integrating with our AI functions.

How it works

It works exactly like our AI chatbots, except it allows you to log in with a username and a password combination. Only after you've logged in, you're able to access it. After you've logged in, you are given access to only AI models that your particular user has access to. This allows a company to create and deploy multiple separate AI models for each department, each model with different functionality and data. You can imagine for instance the marketing department having access to a separate list of AI models than the HR department, etc. Basically ...

The AI Expert System becomes a single UI for all your AI needs

Just like our AI chatbot technology, the AI expert system can execute AI functions, which allows you to trigger AI workflows from within its UI, that starts some sort of workflow that actually does something useful. Imagine the following prompt as an example.

Scrape www.foo.com and create a sales pitch email for why they need our services. Then send the email to John Doe, john@doe.com, and CC me.

The above is of course just a tiny example, and something I'll probably create as an example use case in the future. Watch the video below for a short demonstration of how the system works.

Monetizing for your Custom GPTs

A couple of months ago OpenAI publicly announced they wouldn't allow the general public to monetize their Custom GPTs. This outraged the entire community and created a lot of problems for people having spent huge amounts of time and effort building really good high quality GPTs.

The paradox is that with our AI Expert System, you can actually create equally complex functionality, and in fact you can easily build 1,000x more complex AI chatbots, and you can actually monetize these and sell access to them as you see fit. In addition, your users will not be capped by a maximum number of requests per hour, since you'd be using OpenAI's API and not the ChatGPT UI.

It is also much easier to integrate an AINIRO AI model with custom functionality, such as an API or your database, then it is to integrate a Custom GPT. All in all, this provides you with "a better Custom GPT story than Custom GPT." Basically ...

The AI Expert System is arguably a better Custom GPT than Custom GPT

The future

Notice, the AI Expert System is to be considered experimental BETA software at this point in time. However, this is something we will be focusing a lot on in the future, to stabilise it and improve upon it.

In addition we are considering allowing you to "white label" the system, which allows you to completely hide AINIRO, something that would imply your customers would feel as if they're using a system entirely created and delivered by you.

If you're interested in playing around with it today though, you can find it in the "Plugins" section of your Magic Cloudlet and install it in some few seconds.

Have fun 😊

Thomas Hansen

Thomas Hansen I am the CEO and Founder of AINIRO.IO, Ltd. I am a software developer with more than 25 years of experience. I write about Machine Learning, AI, and how to help organizations adopt said technologies. You can follow me on LinkedIn if you want to read more of what I write.

Published 10. Jun 2024

Creating an AGI Assistant in 22 minutes

In this article I am demonstrating how to create an AGI chatbot in 22 minutes, using Magic Cloud, that allows me to keep track of contacts, scrape websites, and send emails.

Read More

The First AGI Company on Earth

I'm not entirely sure, but I think we are as of now the first AGI company on Earth - And more importantly, we're selling AGI to your company, starting out at $298 per month.

Read More

Software Development is Dead

For decades we've gotten used to a specific process of creating and assembling software. With our AGI framework, this process is now dead and obsolete.

Read More