Cuestionario vinculado a una guía · fácil Nivel

¿El ChatGPT cumple con HIPAA? cuestionario

Prueba cuándo se requiere un acuerdo de socio comercial, qué implica la desidentificación de Safe Harbor y por qué desactivar la capacitación no hace que ChatGPT cumpla con HIPAA.

Rutas de guía relacionadas¿Cumple ChatGPT con Hipaa?
Pregunta 1 de 8

¿Por qué los planes ChatGPT para consumidores, como Free o Plus, no son adecuados para la información de salud protegida?

Vista previa de cada pregunta8 preguntas

Las respuestas y explicaciones aparecen a medida que realiza el cuestionario.

  1. ¿Por qué los planes ChatGPT para consumidores, como Free o Plus, no son adecuados para la información de salud protegida?
    • They encrypt data too heavily
    • OpenAI does not sign a business associate agreement for those plans
    • Solo trabajan en ingles.
    • HIPAA bans all AI tools
  2. Según HIPAA, ¿cuál de ellos es un socio comercial?
    • A patient reading their own chart
    • A hospital's parking contractor that never sees patient data
    • A vendor that creates, receives, stores or transmits PHI on a covered entity's behalf
    • A health plan
  3. Un proveedor anuncia su chatbot como "certificado HIPAA". ¿Qué dice la guía sobre ese reclamo?
    • There is no official government HIPAA certification, so the badge is marketing
    • It means HHS audited the product
    • It replaces the need for a BAA
    • It is required by the Security Rule
  4. Un paciente pega sus propios resultados de laboratorio en ChatGPT. ¿Cómo se aplica HIPAA?
    • The patient has violated HIPAA
    • OpenAI becomes the patient's covered entity
    • The lab must report a breach
    • HIPAA does not restrict what patients do with their own information, but the data then falls under OpenAI's terms
  5. ¿Qué requiere el método de desidentificación Safe Harbor?
    • Removing only the patient's name
    • Encrypting the file
    • Removing 18 kinds of identifiers, including dates more specific than the year
    • Getting the patient's verbal permission
  6. A clinician turns off model training in ChatGPT settings, then pastes a patient note. Why is this still a problem?
    • Turning off training does not create a BAA
    • The setting only works on weekends
    • Training settings delete the account
    • It isn't a problem once training is off
  7. Why does the guide call a BAA 'necessary but not sufficient'?
    • BAAs expire after 30 days
    • The organization still has to do a risk analysis, configure access and retention, and disable features outside the BAA's scope
    • BAAs only cover paper records
    • A BAA must also be approved by the FDA
  8. After someone pastes PHI into an unapproved chatbot, what does the guide say generally has to happen?
    • Nothing, if the chat is deleted
    • The employee must pay a fixed fine
    • The patient must be discharged
    • A breach risk assessment weighing the nature of the PHI, who received it, whether it was viewed and how well the risk was mitigated