Personal AI lets you create AI Personas to organize your knowledge, similar to folders on your computer that you can chat with.
Each AI Persona has access to a set of memory blocks inside your Personal AI that it uses to generate replies. You can train them to excel at customer service, marketing strategy or business operations. In essence, you can treat them like virtual employees with perfect memory on their topics, always ready to answer questions about anything they were previously trained on.
Separating information across AI Personas is key as it controls which kinds of data different users have access to. For example, your customers shouldn’t be able to get answers about your strategy documents.
At the same time, each AI Persona can have a unique focus, configuration and writing voice. This helps you customize the behavior of your virtual employees while replying to your questions, or staying on brand if they’re interacting with people external to the company.
Planning your account with AI Personas
Choosing which AI Personas you’ll train ensures that each receives the data it needs to be useful, while protecting sensitive data from external users.
Here’s how to think about the process of creating and configuring a new AI Persona:
What’s the objective of this persona? Think of how this persona will help you. Will it be a research assistant? A brainstorming buddy? A chatbot to share with others?
What is the target audience? Who is this persona going to be talking to? Only you? Customers? Coworkers?
What kinds of content will you stack inside it? These vary depending on the objective and target audience. For customer service, you may need to stack FAQs and help content. For a marketing strategist persona, you’ll need to stack all the metrics, stats and reports related to marketing.
It’s important to keep these answers clear while creating your AI Personas for a few reasons. First, you’ll keep all information on the same topic and role in the same place, increasing the value of the replies of that persona. At the same time, you’ll be preventing irrelevant or sensitive information from appearing in the conversation, improving data safety. Finally, it’ll help you decide faster which kind of data goes where, making the stacking process more intuitive as you grow your AI in the future.
.svg)
