PIA chat features
Your own PMI assistant!
The PIA chat is currently still in Preview for select properties but is planned to be released in Q2-Q3 2024.
What is PIA chat?
PIA chat is an intelligent digital assistant powered by AI, designed to help you manage your department more efficiently. It provides insights, recommendations, and tools to simplify your workflow.
Enhance Analysis with PIA chat
To analyze your current screen, open the query library and look for items with a camera icon on the “Run this query” button. PIA chat will review the content and provide actionable insights based on the displayed data. You can also take a screenshot using the camera icon at the bottom of the chat interface and ask a question. Additionally, you can upload an image for the PIA chat to analyze.
Can PIA chat assist with finding specific information?
Yes! Use the search feature in the knowledge base tab to type your query. PIA chat will search its semantically indexed database to find the most relevant articles for you.
What if PIA chat cannot resolve my issue?
If PIA chat can’t resolve your issue, you can create a support ticket within the bot interface. The issue will be escalated to a human support agent if necessary.
How does PIA chat handle complex queries?
PIA chat can handle complex queries efficiently by analyzing your screen’s context and providing accurate, relevant responses. For particularly complex issues, it may refer you to a human agent.
Is PIA chat multilingual?
Yes, PIA chat supports multiple languages, ensuring it can assist users from various linguistic backgrounds effectively.
Here’s a visual breakdown of the PIA chat interface:
- PIA chat Avatar Icon: This represents your AI assistant.
- Welcome Message: PIA chat greets you and provides a brief overview of its capabilities.
- Quick Navigation buttons: Easily access key features.
- Query Library: A repository of common queries and solutions.
- Grab Screenshot: Capture the current browser window for context.
- Image Upload button: Attach files directly within the chat.
- Input Text Message: Type your queries or messages here.
- Send Message Icon: Click to send your message. It changes to a stop icon when responding, allowing you to interrupt if needed.
- Knowledge base tab: Search and read articles from the Knowledge Base. The PIA chat will always add footnotes in its responses, directing you to more information on any specific matter or topic.
- References and Citations: When providing an answer, the AI will indicate the primary sources it consulted. These references highlight where the AI retrieved the main information used to formulate the response, allowing users to explore the original content for more context or detailed understanding.
- Good response: Click here if you’re satisfied with the response.
- Bad response: Opens a feedback window so you can explain what was wrong with the prompt.
- Copy Text: Copies the last response to the clipboard for easy sharing or further use.
- Try again: Allows you to request a new response (or retake) on the provided answer.
- Clear chat history: Clearing the chat history resets the chat window and removes the AI’s memory of the conversation. Typically, the AI retains the last five user messages and its corresponding responses, using this history to inform its replies. When you clear the chat history, this context is erased, and the AI will no longer reference those previous interactions in its responses.
For more detailed information and tips on mastering PIA chat, visit Mastering PIA chat for Enhanced Productivity.
Troubleshooting tips
This section provides additional troubleshooting advice for readers experiencing specific issues. While it’s packed with useful insights, you may skip it if you’re not facing any related problems.
Efficient Data Analysis Without Using the Query Library
If you don’t use the query library, you can still take screenshots and ask questions. The bot remembers the last 5 interactions, allowing you to ask it to analyze data in light of recent screenshots. You can switch between views, capturing snapshots and instructing the bot to wait for the next screenshot before responding. This way, you can efficiently gather and analyze data across different screens.