At Pulsora, we are constantly evaluating the latest technology options to empower our customers and provide benefits like increased accuracy, higher efficiency, or better insights. With the rise of artificial intelligence (AI) and machine learning (ML) across areas like generative AI, natural language processing (NLP), and other AI-powered platform capabilities, we at Pulsora are focused on making sure that those capabilities while delivering the benefits first meet the stringent data security and privacy requirements for our customers. In this blog, we share some key aspects of how we have enabled PulsoraAI to meet those requirements.
What is PulsoraAI?
PulsoraAI is a suite of seamlessly integrated, AI-powered capabilities that enhance the features of the Pulsora platform, delivering added value to customers.
What AI, ML and LLMs does PulsoraAI use?
There are two approaches we have taken for PulsoraAI features based on the use cases and the underlying data required;
1. Leverage AI and ML algorithms on the Pulsora cloud platform. This approach is very relevant for use cases where we want to keep customer-specific data on the Pulsora cloud platform and not send it to external LLMs. This also gives us greater customization capabilities that can be leveraged to make the feature more relevant and specific to the Pulsora use case.
2. Leverage Google Gemini as the generative AI LLM. This approach allows us to leverage the existing Google Gemini LLM into the platform with minimal development time. It provides high quality out-of-the-box access to a well-trained model with extensive data coverage.
How does Pulsora AI work?
- Pulsora application receives a request from a user that requires processing using PulsoraAI.
- The request is passed to AI models and LLMs based on the feature. If needed, the user’s request is augmented with data from their organization to generate a search query most relevant for the user request.
- Based on the user request and response the feature requires the platform leverages one of the two paths listed below. The selection of the processing path for each feature is pre-determined by the platform.some text
- Pulsora sends the query to the Google Gemini LLM for processing to generate a response that fulfills the user’s request.
- Pulsora sends the query to the AI Model on the Pulsora platform for processing to generate a response that fulfills the user’s request.
- Pulsora processes the output from the LLMs and AI Models formats and displays the output to the user.
How is my data protected?
Managing customer data security and privacy
On the Pulsora platform, customers own their data and have full control over it. When the data is sent to the AI model or algorithm, it is anonymized, and no customer-identifiable information is sent to the external Google Gemini Generative AI (Gen AI) Large Language Model (LLM).
Pulsora AI honors existing application and data permissions. The LLMs and AI Models used to generate AI responses for a user cannot see or use any information to which that user does not already have access.
User settings and control
While most organizations want to explore PulsoraAI and leverage its benefits, there are different use cases. We have made it an explicit choice for our customers to enable PulsoraAI features that are relevant to them and enable them selectively vs. all-or-nothing enablement of AI capabilities.
Transparency and trust
When users interact with PulsoraAI capabilities, we’ve provided clear visual indicators that inform them about it. This ensures that the user is aware that they are interacting with an AI system and can understand and evaluate those actions before proceeding.
Customer data segregation
Individual customer accounts are kept separate in our production environment. We do not mix or process data from different customers together during AI processing. This means we do not expose your data to other Pulsora customers.
As we all move forward in this new era of AI, we at Pulsora are committed to ensuring that the fundamental principles of data security and privacy are applied to all aspects of the platform ensuring that our customers' data is always safe and in their control.
