Our latest challenge…

Our innate need and urgency for innovation has always driven us to develop solutions aimed at improving our condition and satisfying our partners and customers .
Thanks to the development processes guided by models (Model Driven Development) and the support of generation frameworks (Model To Text) built in recent years, we have achieved remarkable results in terms of productivity and quality of the systems we created. However, reaching these levels has not satisfied our desire for growth and our avid quest for renewal.
Based on these premises, we present our latest professional challenge, aimed at providing assistance to any type of management structure through the use of Machine Learning systems
for the ingestion and processing of company documents, and Artificial Intelligence for the comprehension and processing of tex t and/or voice in natural language.

The system is presented as a chatbot-assistant that allows the user to request data, aggregations, suggestions, and elaborations in the same way we could with a real assistant, i.e. , through natural language. The assistant will take care of finding the most relevant context for the request, based on the company data that will have to be be previously loaded and organized
into a document subsystem that will equip the chatbot.

To improve all of this and make it easier to integrate into business contexts and with different hierarchical and visibility levels, an advanced document classification system comes into play. The
classification will allow the system to respond only based on the contex ts that the user is authorized to consult.
To give an ex ample, in a travel agency, the consultant will have access to all data related to possible travel packages and options, also having the possibility to directly request quotation based
on customer needs . Naturally, they cannot request the total amount of all unpaid invoices (internal to the company) for 2022, as those data have a level of classification above their visibility.

A manager of the same company (with a higher level of visibility) will instead have free access to that data and will be easily able to ask the chatbot to use them for a consultation or to draft an email to the respective debtors . The benefits are therefore evident, and we summarize the main ones below:

·Speed and A ccuracy : an AI- based assistant can process information much faster than humans and perform calculations or data searches with much greater precision.

·Availability : the assistant can be available 24/7 without the need for breaks or days off. This means that users can access assistance whenever they need it, without having to wait for office hours or for a human assistant to be available.

·Learning ability : the assistant can continuously learn from user data and interactions, improving its performance over time. This means that it can become increasingly effective and precise in providing assistance, unlike a human assistant, who may have limitations in their knowledge or skills

·Cost reduction: the assistant can certainly be cheaper than human assistants, as it does not require a salary or health insurance. This means that companies can provide assistance at a lower cost, and users can access more affordable assistance services .

Finally, if we want to find a negative aspect of the system, we cannot ignore the fact that artificial intelligence is still in the early stages of its evolution, and- although its potential is already enormous- there has not yet been enough time to determine its limits and risks:

·Natural language understanding is currently remarkable, but there are still contexts, such as those related to emotions, that can cause the AI to misinterpret the user’s intention due to its lack of implicit empathy.

·One of the risks that most concerns users is related to ethics . For various reasons, AI could prove to be discriminatory or prejudiced in certain decision-making processes .

Our intention is to make the system available in beta soon, allowing interested parties to receive early access to what has been described.

You are all invited to sign up for our waiting list not to miss any updates .