FAQs

• How does Harriet work?

  Harriet is an AI assistant designed to support your team right where they work—on Slack, Teams, or Gchat. Powered by advanced language models, Harriet is trained on your company’s unique information, so she can answer internal questions, guide employees through processes, and streamline tasks effortlessly.

  Because Harriet is connected to your systems and data, she can also automate routine tasks, like submitting tickets, tracking approvals, and providing real-time updates on schedules, policies, and more.

  Get in touch for a full demo.

• Is my data safe?

  Yes, your data is safe with Harriet. We never share private or personal information with the AI models that power Harriet. All personal details are removed from your team’s communications before processing. No sensitive data is stored in our database or sent to the underlying models, so your information will never be used to train AI.

• How is this different from chat GPT?

  We use advanced LLMs like ChatGPT, as well as similar systems from Anthropic, Google, and others, to power Harriet. But unlike generic AI tools, Harriet is designed to work specifically within your organization.

  Harriet connects these powerful LLMs to your internal systems and documents, enabling them to provide accurate, personalized answers while operating safely and securely. Unlike ChatGPT, which pulls information from the internet, Harriet is trained on your policies, processes, and organizational structure.

  This means Harriet can go beyond generic conversations to address strategic needs, like:

  • Providing consistent support across teams and geographies.
  • Enabling employees to self-serve detailed, context-specific information.
  • Automating complex workflows securely, without risking sensitive data.

  Harriet transforms AI into a practical and secure tool for your business.

• Is Harriet anonymous?

  Yes, Harriet allows users to remain anonymous in their interactions, except when compliance or legal reasons require information to be identified. Harriet ensures transparency by confirming with users before escalating any question to HR or sharing their details, so you’re always in control.

• Can Harriet be used globally?

  Yes, Harriet is designed to work seamlessly across geographies. You can upload policies for multiple regions, and Harriet will tailor answers based on where the question is being asked.

  Harriet is also multilingual, able to read information in one language and apply it in another. It supports complex regional and permissions systems, ensuring the right information reaches the right people. Plus, with integrations like Okta, Harriet can fit securely into your global tech stack.

• Can Harriet be used with Microsoft Teams?

  Yes! And Sharepoint too

• Can Harriet be used with Gchat?

  Yes!

• How long does it take to implement Harriet?

  Harriet customers can be live in as little as two weeks, although more complex configurations may require longer.

• Can Harriet provide reporting and insights?

  Yes, Harriet tracks interactions to provide insights into employee needs, workflow bottlenecks, and system performance, helping you identify trends and improve processes.

• How scalable is Harriet?

  Harriet is built to grow with your organization, supporting additional geographies, languages, policies, and systems as needed. Harriet support organizations up to 10,000+ FTE.