A basic understanding of AI terms helps you use Dragon Copilot features effectively. Here are the key terms.
Artificial intelligence (AI): AI enables machines to perform tasks that usually require human intelligence, such as making decisions, translating languages, recognizing patterns, and answering complex questions. It learns from large amounts of data by using algorithms.
AI acts like a smart assistant that helps with tasks by learning, understanding, recognizing, and deciding.
Responsible AI: Responsible AI guides people to design AI systems that are safe, fair, and reliable across models, applications, interfaces, and access controls.
Responsible AI means using AI ethically and making sure it benefits everyone.
Microsoft applies Microsoft's responsible AI practices to all AI solutions.
Conversational AI: Conversational AI helps computers understand, interpret, and respond to human speech. Examples include virtual assistants and chatbots.
Dragon Copilot trains on billions of minutes of clinical narrative and millions of clinical conversations, so it can recognize speech in clinical settings and use it to inform generative AI output.
Generative AI: Generative AI creates new content, such as text and images, by learning patterns from existing data.
Dragon Copilot uses fine-tuned generative AI to create accurate, consistent documentation efficiently. After it captures a clinician-patient conversation, it can generate a clinical note, referral letter, after-visit summary, and other note support.
Ambient AI: Ambient technology fits naturally into your workflow and works in the background.
In Dragon Copilot, the app securely captures the clinician-patient conversation in the background. This lets you focus on patient care while Dragon Copilot turns the conversation into high-quality, specialty-specific notes.
Copilots: A copilot is an AI-powered assistant that helps with tasks such as writing, coding, summarizing, and searching across apps. It understands natural language and can answer questions, create content, or take action while you work.
Developers build copilots with Responsible AI guardrails to keep them safe and secure.
Like an airplane copilot, it helps you work more efficiently, but you stay in control.
Prompts: A prompt is an instruction in language, images, or code that tells AI what to do.
You can use prompts in Dragon Copilot to guide results. For example, verbalization best practices and specialty-specific models help the software place content in the correct section of a summary.
Microsoft is built on trust.
Microsoft protects data and builds every AI solution on a foundation of privacy.
- Uses the Microsoft Secure Future Initiative to support high standards for security, privacy, and compliance.
- Follows Microsoft's Responsible AI practices and applies healthcare-specific clinical, chat, and compliance safeguards to support safe, accurate AI output.
- Builds on privacy principles with transparent policies and strong protections.
Microsoft continues to advance responsible generative AI and help healthcare organizations improve experiences and outcomes across health and life sciences.