Appian AI Copilot empowers all users to collaborate with generative AI to improve efficiency throughout the Appian platform.
With AI Copilot:
This page describes AI Copilot's powerful capabilities, and how it enables you and end users to take advantage of those capabilities.
This feature requires additional configuration in the Admin Console before you can use it.
AI Copilot allows you to harness the power of generative AI to help you quickly build interfaces that provide a beautiful and effortless experience for your users. You can use AI Copilot to generate an interface directly from a PDF in just a few clicks.
Generating a beautiful form that users will love is as easy as 1, 2, 3:
After the form is generated, you can take over. Verify that the generated content is correct and appropriate, then personalize it using our intuitive, low-code design tools.
For the best results, the PDFs you upload should be forms that users can input text into. Flattened PDFs are not supported.
Specifically, AI Copilot supports the following types of PDFs:
Sometimes AI Copilot generates incorrect SAIL when it is generating a form. If this happens, it will stop generating, allowing you to take over. You can try to upload the same form again, but if it keeps having issues you may want to consider modifying the form or using a different one.
If the PDF you upload is too large, AI Copilot will let you know that it couldn't generate the entire form. The size of PDF that AI Copilot can comfortably work with is determined by the token limit of your Azure OpenAI Service license.
First, make sure your PDF is a supported type. Also, make sure the form is human readable. If a human can't read it, AI Copilot can't read it either.
Sometimes, you'll see this error when a PDF isn't created correctly. For example, if a document is renamed using the .pdf extension, but it isn't saved as a PDF.
If your Azure OpenAI credentials are not correct, AI Copilot will display, but it will be unable to generate a form. Make sure the credentials in the Admin Console are correct and try again.
AI Copilot will ask if you want to generate instructions. If you click Yes, it generates instructions in a card next to the form fields. These instructions are based on the content of your form and generally include a summary of the form and instructions for how to fill out the form.
If your form has text that AI Copilot recognizes as instructions, it may add that text to the header of the form.
If you don't select Yes or No, instructions won't be generated. If you'd like to generate instructions after all, click Start over to upload your form again and select Yes the next time AI Copilot asks.
As always, be sure to review any AI-generated content to make sure it is accurate and appropriate.
When AI Copilot is generating your form, we disable some parts of the interface object, such as expression mode and the Preview and Performance tabs. This makes sure that AI Copilot isn't interrupted while it is generating the form. As soon as the form generation completes, the parts of the interface object that were disabled will be reenabled.
The Records Chat component brings the power of AI Copilot to records and their related record data. Add this input field to any interface to empower users to ask questions about a record using a conversational AI chatbot. For example, if you use Appian for case management, you can use this component to synthesize case information, understand its history, and suggest or summarize next steps.
Your users won't need to engineer the perfect prompt to get the insights they're looking for. Users can chat in their own words, and AI Copilot understands what they need. User inputs and chatbot outputs are all protected with private AI; no third-party integrations needed to get started.
The Records Chat component is powered by private AI, meaning that your inputs and outputs are protected and remain within the boundaries of your Appian Cloud site at all times.
Attention Appian Cloud FedRAMP, HIPAA, or PCI-DSS customers: Use of the Records Chat component results in sending data outside of Appian Cloud's FedRAMP, HIPAA, or PCI-DSS in-scope service boundaries. Customers should use caution when using this component with regulated data, and review for their own compliance requirements. Before enabling this feature, please review its compliance to ensure it aligns with your organization's security requirements.
This component is available only on Appian Cloud. Customers will need to contact Appian Support to enable this feature. You will receive the following error until the feature is enabled on your site: This function is currently unavailable. Please contact your system administrator if this problem persists. Self-managed and Appian Government Cloud customers don't have access to this feature.
Additionally, this feature is available for environments in these regions:
To get access to this feature, customers should contact Appian Support.
This feature requires additional configuration in the Admin Console before you can use it.
AI Copilot empowers any user to gain real-time insights about their data. When business users leverage Appian's self-service analytics capabilities, they can partner with AI Copilot to uncover new patterns, trends, or outliers in their data.
Users can ask questions about their reports using a familiar chat experience. To help them get started, AI Copilot provides a list of suggested questions to uncover potential areas of improvements and next steps. AI Copilot will then analyze the report data and provide users with immediate answers.
No. AI Copilot can only assist users in understanding their custom-built reports. With Appian's self-service analytics capabilities, users have the freedom and flexibility to explore enterprise data. AI Copilot helps users dig deeper into their reports so they can analyze the results faster.
If AI Copilot cannot analyze a report, it may be because:
A session refers to a conversation between a user and AI Copilot. During a session, a user can ask AI Copilot questions about their data. At some point, a user may receive a message from AI Copilot indicating that the conversation has reached its limit. This occurs when the user exceeds the token limit on a given session.
The number of questions a user can ask AI Copilot in a given session is determined by the token limit of your Azure OpenAI Service license.
If a user hits the limit, they will need to start a new session to continue asking questions about their report. Users will either be promoted to start a new session, or they can click Restart your conversation with AI Copilot to manually start a new session.
No. If a user starts a new session, AI Copilot will not have any context about the previous conversation. This means AI Copilot won't have answers to follow-up questions from a previous session. Users will need to provide AI Copilot with the necessary context at the beginning of each session.
You don't need an Azure OpenAI license to use the Records Chat component.
Appian fits within your enterprise ecosystem, seamlessly integrating with Azure OpenAI Service.
If you have a license for Azure OpenAI Service, you can use it to enable AI Copilot. Simply provide your Azure OpenAI Service credentials in the Admin Console to enable AI Copilot. After credentials are provided, AI Copilot capabilities will display in your environment.
When you or an end user interact with AI Copilot, certain information is sent to Azure OpenAI to process the request:
| Action | Information sent to Azure OpenAI |
|---|---|
| You upload a PDF to generate an interface. | The content of the PDF. |
| A user asks a question about a their custom-built report. | The question and the selected report data. |
For more information about Azure OpenAI data privacy, see the Microsoft documentation.
