How does it work?

Healthcare Practitioners

Radiologists

50.7%Moderate Risk

Summary

Radiology faces moderate risk as AI excels at routine image analysis, data synthesis, and report drafting. While computer vision can identify anomalies and automate documentation, it cannot replace the physical dexterity required for interventional procedures or the empathy needed for sensitive patient counseling. The role will shift from primary image reading toward high-level clinical consultation and complex, hands-on medical interventions.

Scored by Gemini 3.1 Pro·How does scoring work?

The AI Jury

ClaudeToo Low

The Diplomat

The core task, image interpretation, scores 75-85% risk, and AI radiology tools are already FDA-cleared and outperforming humans on specific tasks. The 50.7% score is anchored too heavily on procedural and interpersonal tasks that represent a shrinking fraction of the job's future.

68%
GrokToo Low

The Chaos Agent

Radiologists patting themselves on the back at 50%? AI's already outdiagnosing you on scans; denial won't save your jobs.

82%
DeepSeekToo High

The Contrarian

Radiologists will become AI conductors, not casualties; automation ignores the legal and clinical nuances that demand human oversight.

42%
ChatGPTToo High

The Optimist

AI will read more scans, faster, but radiologists still own the gray areas, procedures, and clinical judgment. This job is changing shape, not vanishing.

44%

Task-by-Task Breakdown

Review or transmit images and information using picture archiving or communications systems.
95

Transmitting and routing digital images through PACS is a highly structured, routine digital task that is already largely automated.

Prepare comprehensive interpretive reports of findings.
85

Multimodal AI and LLMs can automatically generate structured, comprehensive draft reports from imaging findings for human review.

Document the performance, interpretation, or outcomes of all procedures performed.
85

Automated medical scribes and EHR-integrated AI tools can seamlessly document procedures and outcomes with minimal human input.

Obtain patients' histories from electronic records, patient interviews, dictated reports, or by communicating with referring clinicians.
80

AI systems can rapidly extract, synthesize, and summarize relevant patient histories from vast electronic health records and previous reports.

Calculate, measure, or prepare radioisotope dosages.
80

Calculating and measuring dosages is a highly structured, rule-based task that automated dispensing systems and software calculators can perform with high precision.

Perform or interpret the outcomes of diagnostic imaging procedures including magnetic resonance imaging (MRI), computer tomography (CT), positron emission tomography (PET), nuclear cardiology treadmill studies, mammography, or ultrasound.
75

Computer vision models are highly capable of detecting anomalies in medical images, acting as a primary reader or highly advanced assistant, though human sign-off remains necessary for complex diagnoses.

Check and approve the quality of diagnostic images before patients are discharged.
75

AI computer vision models can instantly evaluate images for technical adequacy, positioning, and completeness, significantly automating quality checks.

Review procedure requests and patients' medical histories to determine applicability of procedures and radioisotopes to be used.
75

AI excels at cross-referencing patient medical histories with clinical guidelines to determine the appropriateness of requested procedures.

Develop or monitor procedures to ensure adequate quality control of images.
70

AI computer vision tools can automatically monitor and flag image quality issues, though human oversight is needed to design the protocols.

Compare nuclear medicine procedures with other types of procedures, such as computed tomography, ultrasonography, nuclear magnetic resonance imaging, and angiography.
65

AI systems can rapidly cross-reference clinical guidelines and patient histories to recommend and compare the efficacy of different imaging modalities.

Test dosage evaluation instruments and survey meters to ensure they are operating properly.
65

Modern medical instruments increasingly feature automated self-diagnostic and calibration software, reducing the need for manual testing.

Prescribe radionuclides and dosages to be administered to individual patients.
60

AI can accurately calculate and recommend precise dosages based on patient data, though a physician must legally authorize the prescription.

Develop treatment plans for radiology patients.
55

AI can synthesize patient data to recommend treatment pathways, but the final high-stakes clinical decision requires human judgment.

Coordinate radiological services with other medical activities.
50

AI can optimize scheduling and workflow logistics, but coordinating across departments requires human negotiation and strategic alignment.

Monitor handling of radioactive materials to ensure that established procedures are followed.
50

AI-enabled computer vision can monitor physical compliance with safety protocols, but human intervention is required to correct and enforce behavior.

Advise other physicians of the clinical indications, limitations, assessments, or risks of diagnostic and therapeutic applications of radioactive materials.
45

Advising peers involves nuanced clinical context, weighing complex risks, and collaborative decision-making that goes beyond simple data retrieval.

Participate in quality improvement activities including discussions of areas where risk of error is high.
40

While AI can analyze data to find error patterns, discussing and implementing systemic quality improvements requires collaborative human judgment.

Direct nuclear medicine technologists or technicians regarding desired dosages, techniques, positions, and projections.
40

While AI can generate the optimal technical parameters, directing and managing human technicians requires interpersonal communication and leadership.

Confer with medical professionals regarding image-based diagnoses.
35

Collaborative clinical decision-making and peer consultations require complex medical reasoning and interpersonal trust that AI cannot replace.

Formulate plans and procedures for nuclear medicine departments.
35

Formulating departmental plans involves strategic foresight, resource allocation, and leadership that AI cannot independently execute.

Communicate examination results or diagnostic information to referring physicians, patients, or families.
30

While AI can draft communication summaries, delivering sensitive diagnostic results requires human empathy, trust, and real-time clinical judgment.

Establish or enforce standards for protection of patients or personnel.
30

Establishing safety standards and enforcing them among staff requires leadership, physical environment awareness, and human management.

Instruct radiologic staff in desired techniques, positions, or projections.
30

Instructing staff on physical positioning and imaging techniques requires hands-on demonstration and interpersonal communication.

Establish and enforce radiation protection standards for patients and staff.
30

Establishing and enforcing safety standards requires leadership, physical environment awareness, and human management.

Provide counseling to radiologic patients to explain the processes, risks, benefits, or alternative treatments.
25

Counseling patients involves empathy, assessing their comprehension, and building trust, which are deeply human interpersonal skills.

Teach nuclear medicine, diagnostic radiology, or other specialties at graduate educational level.
25

Graduate-level medical teaching requires deep mentorship, assessing student comprehension, and complex interpersonal communication.

Administer radioisotopes to clinical patients or research subjects.
20

The physical administration of radioactive materials requires strict safety protocols, hands-on patient care, and precise physical handling.

Recognize or treat complications during and after procedures, including blood pressure problems, pain, oversedation, or bleeding.
10

Managing acute procedural complications requires real-time physical intervention, rapid clinical judgment, and hands-on patient care.

Participate in continuing education activities to maintain and develop expertise.
10

Maintaining personal medical expertise is an inherently human cognitive process that cannot be outsourced to a machine.

Perform interventional procedures such as image-guided biopsy, percutaneous transluminal angioplasty, transhepatic biliary drainage, or nephrostomy catheter placement.
5

Interventional procedures require extreme physical dexterity, real-time spatial adaptation, and high-stakes physical manipulation that robotics cannot fully automate in the near term.