Ai in Healthcare: What Patients Need to Know

• A. Robots that perform surgery without any human involvement
• B. Computer systems that perform tasks that normally require human intelligence, such as recognizing patterns or making predictions ✓
• C. Electronic health records that store patient information
• D. Medical devices that only measure vital signs like heart rate and blood pressure

Why: AI refers to computer systems that can perform tasks typically requiring human intelligence, such as analyzing images, identifying patterns in data, or predicting outcomes. Surgical robots, health records, and vital sign monitors may be part of healthcare technology, but they are not AI by definition.

• A. Replacing all doctors and nurses with automated systems
• B. Analyzing medical images such as X-rays or scans to help identify possible signs of disease ✓
• C. Automatically prescribing medications without a clinician's review
• D. Deleting patient records after a set period of time

Why: One of the most established uses of AI is helping clinicians analyze medical images to flag areas of concern. In most settings, AI supports rather than replaces clinicians, who make the final decisions about diagnosis and treatment.

• A. So they can avoid all healthcare technology
• B. So they can ask questions, understand how decisions are made, and give informed consent ✓
• C. Because AI always makes mistakes and should be ignored
• D. Because patients are legally required to operate AI tools themselves

Why: Transparency helps patients take an active role in their care. Knowing when AI contributes to a decision allows patients to ask questions, understand the reasoning, and make informed choices about their treatment.

• A. The risk that AI will always favor the most expensive treatment
• B. The risk that an AI system trained on unrepresentative data may perform less accurately for certain groups of patients ✓
• C. The risk that patients will be charged more for using AI tools
• D. The risk that AI will refuse to answer any medical questions

Why: AI learns from the data it is trained on. If that data underrepresents certain ages, ethnicities, sexes, or conditions, the system may be less accurate for those groups, which can lead to unequal care.

• A. AI results should be accepted without any review by a clinician
• B. AI can provide useful information, but a qualified healthcare professional should review it and make the final decision ✓
• C. AI is only used by researchers and never affects individual patients
• D. AI diagnoses are always more reliable than any human doctor's judgment

Why: AI can be a valuable tool for spotting patterns and supporting decisions, but it has limitations and can make errors. A qualified clinician should interpret AI output in the context of the whole patient, which is why human oversight remains essential.

• A. So the hospital can charge a higher fee for AI-assisted care
• B. So patients can make informed decisions about their care and understand how their information is being used
• C. So the AI system can learn faster from each patient's reaction
• D. So doctors can avoid responsibility for their decisions

• A. The AI system runs slower on older computers
• B. The AI produces systematically unfair or inaccurate results for certain groups, often because of unrepresentative training data
• C. The AI always recommends the most expensive treatment
• D. The AI refuses to answer questions about medication

• A. AI systems often require large amounts of sensitive health data, which can be misused or exposed if not properly protected
• B. AI systems cannot store any data at all
• C. Patients are legally barred from viewing their own medical records when AI is involved
• D. Privacy only matters for surgical procedures, not for AI tools

• A. No one, because AI cannot make mistakes
• B. Only the patient, for not questioning the result
• C. The healthcare providers and organizations that deployed and relied on the tool, since human oversight remains essential
• D. Only the AI itself, which can be fined

• A. Patients agreeing to a hospital's general terms of service without reading them
• B. Patients being given clear, understandable information about how AI will be used in their care, including its benefits and limitations, before agreeing to it
• C. Doctors informing the AI system about the patient's medical history
• D. A signed form stating the patient will never question the AI's recommendations

• A. AI systems never need access to patient data to be useful
• B. AI systems often require large amounts of data, so patients should understand how their data is collected, stored, and used
• C. Patient data is always fully anonymized before AI systems see it
• D. Once data is given to an AI tool, it cannot be shared or reused

• A. It allows hospitals to sell patient data without restriction
• B. It ensures patients understand how their data will be used by AI and can agree or decline before it happens
• C. It replaces the need for doctors to explain treatment options
• D. It guarantees that AI recommendations are always accurate

• A. Deleting all medical records permanently
• B. Removing or masking information that directly identifies a person, such as name or address, before data is used
• C. Encrypting data so only the patient can read it
• D. Giving patients a new identity in the healthcare system

• A. Patients are legally required to operate AI tools themselves
• B. Asking questions helps patients understand how their data is protected and how AI influences their care decisions
• C. Questions automatically disable the AI system
• D. Hospitals only accept patients who ask about AI

• A. Sharing login passwords with family members for convenience
• B. Using strong, unique passwords and reviewing privacy settings and consent options in health apps
• C. Posting screenshots of AI-generated health results on social media
• D. Accepting all data permission requests without reading them

• A. To replace doctors entirely in making final diagnoses
• B. To analyze medical data such as images and test results to help clinicians identify possible conditions ✓
• C. To perform surgery without human involvement
• D. To schedule patient appointments automatically

Why: AI in diagnosis primarily supports clinicians by analyzing data such as X-rays, scans, and lab results to flag patterns that may indicate a condition. Doctors remain responsible for interpreting results and making final decisions.

• A. Patient billing records
• B. Medical images such as X-rays, CT scans, and MRIs ✓
• C. Hospital parking logs
• D. Staff email communications

Why: Medical imaging is one of the most common applications of AI in diagnosis, as algorithms can be trained to detect patterns like tumors, fractures, or signs of disease in images.

• A. Because AI results are always wrong
• B. Because doctors can consider the full context of a patient's health, history, and symptoms that an AI tool may not fully capture ✓
• C. Because AI tools are illegal to use in healthcare
• D. Because patients are not allowed to see their results

Why: AI tools analyze specific data inputs and may miss context such as personal medical history, lifestyle, or symptoms not recorded. A doctor can combine AI output with this broader picture to make a sound judgment.

• A. A correct diagnosis of a serious illness
• B. A result indicating a condition is present when it actually is not ✓
• C. A test that the patient cancelled
• D. A diagnosis confirmed by two doctors

Why: A false positive means the AI flags a condition that is not actually there. Understanding this helps patients realize why follow-up testing and clinical review are important before acting on a result.

• A. It guarantees a cure for every disease
• B. It can help detect certain conditions earlier and support faster, more consistent analysis ✓
• C. It eliminates the need for any medical tests
• D. It removes all costs from healthcare

Why: AI can process large amounts of data quickly and consistently, which may help identify some conditions earlier. However, it does not guarantee cures or eliminate the need for testing and professional care.

• A. To give every patient the same treatment for a given disease
• B. To tailor prevention and treatment to an individual's genetic, lifestyle, and environmental characteristics ✓
• C. To eliminate the need for doctors by using only automated systems
• D. To reduce the cost of hospital stays by shortening appointments

Why: Personalized medicine uses information about a person's genes, lifestyle, and environment to choose the most suitable prevention or treatment, rather than a one-size-fits-all approach. AI helps analyze these complex data to support those choices.

• A. By replacing the patient's medical history with random guesses
• B. By analyzing large amounts of patient data to identify patterns that suggest which treatments may work best for a specific person ✓
• C. By automatically prescribing medications without any clinician involvement
• D. By storing patient data without ever analyzing it

Why: AI can process large datasets, such as genomic information, lab results, and health records, to find patterns that help predict how an individual may respond to a treatment. Clinicians still make the final decisions.

• A. The study of how medicines are packaged and shipped
• B. The study of how a person's genes affect their response to medications ✓
• C. The study of hospital building design
• D. The study of how AI systems are programmed

Why: Pharmacogenomics examines how genetic differences influence how people metabolize and respond to drugs. AI can help identify these patterns, allowing doctors to choose medications and doses that are more likely to be safe and effective.

• A. Health data should always be shared publicly to improve AI accuracy
• B. Patients should understand how their data is collected, stored, and used, and have appropriate consent and protections ✓
• C. Privacy does not matter once data is converted into numbers
• D. Only doctors, never patients, should know how their data is used

Why: Because personalized medicine relies on sensitive personal data, patients should know how their information is used and protected. Informed consent, secure storage, and clear policies help maintain trust and protect patient rights.

• A. Because AI recommendations are never accurate under any circumstances
• B. Because doctors can consider the full context of a patient's health, preferences, and values, and can explain and check AI suggestions ✓
• C. Because patients are legally required to memorize all AI outputs
• D. Because AI systems only work when a doctor is absent

Why: AI can support decisions, but it may miss context or make errors. A clinician can evaluate recommendations alongside the patient's full situation, explain the reasoning, and make a shared decision with the patient.

• A. They never make mistakes when trained on large datasets
• B. They can produce errors when the patient data differs from the data they were trained on ✓
• C. They can fully replace the judgment of a trained clinician
• D. They are unaffected by the quality of the information they receive

Why: AI systems learn patterns from their training data. When a patient's situation or population differs substantially from that data, the tool may perform poorly, which is why clinical oversight remains essential.

• A. AI tools always perform equally well for all patients
• B. The training data may underrepresent certain ages, ethnicities, or genders ✓
• C. AI tools only work for patients who live in cities
• D. Patients over 40 cannot be analyzed by AI systems

Why: If certain groups are underrepresented in the data used to build an AI model, the model may be less accurate for those groups. This is a known source of bias in healthcare AI.

• A. It is stored in a physical container in the hospital
• B. Its internal reasoning is difficult for humans to understand or explain ✓
• C. It only works when the lights are off
• D. It can only be used by a single doctor

Why: Many AI models, especially complex deep learning systems, do not clearly show how they reached a conclusion. This lack of explainability can make it harder for clinicians to verify or question the output.

• A. AI results are always wrong
• B. A doctor can consider the full context of your health history, symptoms, and circumstances ✓
• C. Doctors are required by law to ignore AI results
• D. AI tools cannot store any patient information

Why: Clinicians combine test results with medical history, physical examination, and other factors. AI outputs are one input among many and should be interpreted within that broader picture.

• A. A preference for machines that run automatically overnight
• B. The tendency for people to trust computer outputs too readily, even when they are wrong ✓
• C. A rule that AI must always be turned off during surgery
• D. The practice of training AI using only automated data entry

Why: Automation bias occurs when clinicians or patients over-rely on AI recommendations and fail to notice or question errors. Recognizing this risk helps keep human judgment active in care decisions.

• A. The right to have the AI system's source code published
• B. The right to be informed that AI is being used in their care ✓
• C. The right to receive a discount on services that use AI
• D. The right to choose which AI company the hospital contracts with

Why: Patients generally have the right to know when AI plays a role in their diagnosis, treatment, or care decisions, so they can ask questions and make informed choices. The other options are not standard patient rights.

• A. The AI software developer alone
• B. The patient, because they agreed to the care
• C. The healthcare provider who makes and communicates the final decision ✓
• D. No one, because AI decisions cannot be assigned responsibility

Why: AI tools are typically decision-support aids. The licensed clinician remains accountable for the final recommendation and its communication to the patient.

• A. Patients must sign a document saying they understand all computer science concepts
• B. Patients receive relevant information about how AI is used and can agree or decline before it affects their care ✓
• C. Patients give the hospital permission to sell their data to AI companies
• D. Patients consent only after the AI has already produced a diagnosis

Why: Informed consent means patients receive understandable information about the role of AI, its benefits and limitations, and alternatives, allowing them to make a voluntary choice.

• A. Does the AI have a logo I can see?
• B. What is the AI's role, and can a human review its results? ✓
• C. Can the AI write my insurance claim for me?
• D. Which AI is the most popular on the market?

Why: Asking about the AI's role and whether a human reviews its output helps patients understand oversight and accountability, which are central to safe and rights-respecting AI use.

• A. AI systems do not store any data, so privacy is never a concern
• B. AI systems often require large amounts of health data, which must be protected under privacy laws and rules ✓
• C. Privacy only matters for paper records, not digital tools
• D. Patients lose privacy only if they request an AI second opinion

Why: Many AI tools rely on large volumes of sensitive health information, so patients have a right to expect that their data is collected, used, and shared lawfully and securely.