Changelog

We're excited to announce our beta release of Extractive AI Answers!

Unlike traditional generative AI models that create every part of the answer, Extractive AI aims to find the best parts from your existing content and uses those to build the response. The main advantage here is reliability: With Extractive AI, we can guarantee to deliver an answer for every customer query - a promise that generative AI models are unable to do.

As it relies primarily on your existing content, Extractive AI also creates an opportunity to refine and bolster your knowledge base - the better your content, the better the customer's experience.

What is Extractive AI?

Unlike traditional generative AI models that attempt to create entirely new content based on patterns and data, Extractive AI focuses on identifying and utilizing the best pieces of information from your existing repository of content to formulate reliable and precise responses. This approach is particularly advantageous for several reasons:

1. Reliability & Continuity: Generative AI may often stumble when it comes to accuracy and coherence, sometimes crafting answers that are plausible but incorrect. With Extractive AI, we significantly reduce this risk by relying on already vetted and verified content from your knowledge base. By doing so, we can confidently deliver an answer to every customer query, ensuring a consistently reliable customer experience.  

2. User Experience: Extractive AI underscores the importance of having high-quality, thorough, and up-to-date content. The benefit of this offering is that Brainfish customers still get all of the other benefits including AI actions, intent identification, analytics and themes.

Industries that Benefit

Our Extractive AI Answers technology is useful across various industries that prioritize accurate and timely customer support, including but not limited to:

1. Finance/Financial Services

In the financial industry, where exact information and adherence to regulations are paramount, Extractive AI can help deliver precise responses regarding interest rates, loan details, fee structures, and more. Ensuring that clients and customers receive the correct information can enhance trust and satisfaction.

2. Healthcare

Healthcare providers managing vast databases of medical information, patient queries, and appointments can utilize Extractive AI to offer accurate medical responses based on existing health records, policy details, and procedural information, ensuring patients receive trustworthy and reliable information without delay.

3. Education

Educational institutions can benefit greatly by offering quick and accurate responses to queries about courses, application processes, deadlines, and other academic-related information. Having a reliable AI-assistant ensures that students and prospective applicants receive accurate and helpful guidance.

Internal Zendesk Content Import: Brainfish now supports importing internal Zendesk content right into Brainfish. This function significantly streamlines the content transfer process, making it easier for Zendesk users to switch to Brainfish.

General Improvements: As part of our ongoing commitment to improve user experience, several tweaks have been made across the board. You may notice some subtle changes in our user interface that make Brainfish more intuitive and user-friendly. As always, if you have any issues or suggestions, don't hesitate to reach out to us at support@brainfi.sh.

In a step towards increasing the precision of our AI, we've quietly rolled out the Intent Classification feature in the backend. This new feature enables Brainfish to not only understand the customer's query but also interpret the underlying intent of the question. Understanding this can help customers in future to customise their responses to cater to the specific needs or concerns raised by the customer.

Widget Behaviour Exposure to DOM

We've further exposed our widget behaviour as Javascript events to provide our customers with enhanced flexibility and control. This means customers can now trigger the widget based off JavaScript events, as well as initiate events based on the closure of the widget.

This improved accessibility and control should allow power users more leeway in customising their widget's implementation, creating a richer, more personalised user experience.

Check out the Help article on the widget here.