Skip to main content
Menu

JOB REPORT / 15-2051 · 54 TASKS

Data Scientists.

Can AI do the work of Data Scientists?

Mostly it works alongside you. Here is what shifts, task by task.

YOUR EXPERIENCE MATTERS

How does this fit your work?

How many tasks do you use AI to assist with at work?

Drag the dot. There’s no wrong answer; it starts at our estimate.

Starting pointOur estimate

The mark shows what’s typical for this job.

The parts of your job AI can already do most of:

  • Analyze, manipulate, or process large sets of data using…
  • Create graphs, charts, or other visualizations to convey…
  • Identify relationships and trends or any factors that could…

Your free lesson plan builds the skills that keep you ahead of AI.

A free set of short lessons, about 20 minutes each, at your own pace. No payment, no experience needed.

TURN THIS INTO SKILLS YOU CAN PROVE Early access

Earn a certificate in your job

A guided certificate program in AI-assisted Data Scientists work. It covers the 54 AI-affected parts of your job and ends in a capstone a person reviews.

WHAT WE LOOKED AT

Your job, task by task.

54 of 54 tasks reviewed

We went through all 54 core tasks in this job, one at a time.

Task by task, from the work Data Scientists actually do. How we describe AI use in each task is our own read of it, not a measurement of whether AI can replace the job.

  1. TASK 01

    Analyze, manipulate, or process large sets of data using statistical software.

    AI DOES

    Writes statistical code, suggests models, explains output, and flags data issues

    YOU DO

    Choose the right question to ask and judge whether results make real-world sense

  2. TASK 02

    Create graphs, charts, or other visualizations to convey the results of data analysis using specialized software.

    AI DOES

    Suggests chart types, writes plotting code, and drafts layouts from data

    YOU DO

    Decide which insight matters most and design visuals that land with your specific audience

  3. TASK 03

    Identify relationships and trends or any factors that could affect the results of research.

    AI DOES

    Scans data for correlations, outliers, and trends using statistical models and visualizations

    YOU DO

    Judge which patterns are meaningful, not just mathematical, and spot confounding factors AI might miss

  4. TASK 04

    Write new functions or applications in programming languages to conduct analyses.

    AI DOES

    Drafts functions, suggests algorithms, and auto-completes code blocks

    YOU DO

    Decide what question to answer and verify the code does it correctly

  5. TASK 05

    Identify or monitor current and potential customers, using business intelligence tools.

    AI DOES

    Scans customer data, finds patterns, scores leads, and surfaces trends automatically

    YOU DO

    Judge which insights actually matter for your specific business goals

  6. TASK 06

    Apply feature selection algorithms to models predicting outcomes of interest, such as sales, attrition, and healthcare use.

    AI DOES

    Runs filter, wrapper, and embedded feature selection methods and ranks variables by importance

    YOU DO

    Decide which features are meaningful for the business problem and flag ones that look good statistically but don't make real-world sense

  7. TASK 07

    Clean and manipulate raw data using statistical software.

    AI DOES

    Writes cleaning scripts, flags outliers, and suggests transformations

    YOU DO

    Decide what 'clean' means for your specific problem and data context

  8. TASK 08

    Compare models using statistical performance metrics, such as loss functions or proportion of explained variance.

    AI DOES

    Calculates loss functions, R-squared, and other metrics, then summarizes model comparisons in tables or charts

    YOU DO

    Choose which metrics fit the real-world goal and judge whether the numbers actually mean the model is ready to use

  9. TASK 09

    Design surveys, opinion polls, or other instruments to collect data.

    AI DOES

    Drafts survey questions, suggests answer scales, and flags common bias patterns

    YOU DO

    Match questions to real research goals and know your specific audience

  10. TASK 10

    Read scientific articles, conference papers, or other sources of research to identify emerging analytic trends and technologies.

    AI DOES

    Summarizes papers, extracts key themes, and flags recurring new methods across dozens of sources fast

    YOU DO

    Decide which trends are worth pursuing based on your team's real needs and context

  11. TASK 11

    Test, validate, and reformulate models to ensure accurate prediction of outcomes of interest.

    AI DOES

    Runs validation checks, generates test metrics, and suggests reformulations automatically

    YOU DO

    Judge whether model logic fits the real business question and catch hidden biases

  12. TASK 12

    Generate data queries, based on validation checks or errors and omissions identified during data entry, to resolve identified problems.

    AI DOES

    Scans datasets, detects rule violations, and drafts standard query language

    YOU DO

    Decide which errors matter clinically and approve queries before they go to sites

  13. TASK 13

    Document specifications for business intelligence or information technology reports, dashboards, or other outputs.

    AI DOES

    Drafts report specs, suggests field definitions, and formats documentation templates

    YOU DO

    Confirm the specs match real business goals and stakeholder expectations

  14. TASK 14

    Write work instruction manuals, data capture guidelines, or standard operating procedures.

    AI DOES

    Drafts templates, organizes sections, and suggests standard regulatory language

    YOU DO

    Verify clinical accuracy and tailor procedures to your specific study and regulations

  15. TASK 15

    Apply sampling techniques to determine groups to be surveyed or use complete enumeration methods.

    AI DOES

    Runs stratified, random, and cluster sampling calculations and flags trade-offs between methods

    YOU DO

    Decide which sampling approach fits the real-world context and research goals

  16. TASK 16

    Identify business problems or management objectives that can be addressed through data analysis.

    AI DOES

    Surfaces common problem types, suggests relevant frameworks, and drafts problem statements from prompts

    YOU DO

    Talk to stakeholders, read the room, and judge which problems actually matter to the business

  17. TASK 17

    Identify solutions to business problems, such as budgeting, staffing, and marketing decisions, using the results of data analysis.

    AI DOES

    Surfaces patterns, generates solution options, and summarizes data insights quickly

    YOU DO

    Weigh trade-offs and tailor recommendations to your company's real-world constraints

  18. TASK 18

    Propose solutions in engineering, the sciences, and other fields using mathematical theories and techniques.

    AI DOES

    Suggests relevant mathematical techniques, generates solution drafts, and surfaces research or prior work fast

    YOU DO

    Judge which solution actually fits the problem's real constraints, risks, and goals

  19. TASK 19

    Recommend data-driven solutions to key stakeholders.

    AI DOES

    Generates draft summaries, visualizations, and talking points from data

    YOU DO

    Read the audience, handle pushback, and connect data to business priorities

  20. TASK 20

    Maintain or update business intelligence tools, databases, dashboards, systems, or methods.

    AI DOES

    Drafts SQL queries, suggests dashboard layouts, auto-completes code, and flags data anomalies

    YOU DO

    Decide what metrics matter, validate data logic, and align tools to real business goals

  21. TASK 21

    Manage timely flow of business intelligence information to users.

    AI DOES

    Schedules report delivery, flags anomalies, and routes data summaries to the right people

    YOU DO

    Decide which insights are urgent and tailor communication to each user's needs

  22. TASK 22

    Perform quality control audits to ensure accuracy, completeness, or proper usage of clinical systems and data.

    AI DOES

    Scans datasets for missing values, outliers, and rule violations — then flags them for review

    YOU DO

    Judge which errors matter clinically and decide how to fix or escalate them

  23. TASK 23

    Conduct or coordinate tests to ensure that intelligence is consistent with defined needs.

    AI DOES

    Runs validation scripts, checks data consistency, and flags mismatches automatically

    YOU DO

    Decide if test results truly match what the business actually needs

  24. TASK 24

    Track the flow of work forms, including in-house data flow or electronic forms transfer.

    AI DOES

    Monitors form movement, flags delays, and generates tracking reports automatically

    YOU DO

    Decide how to handle exceptions, compliance issues, and cross-team communication

  25. TASK 25

    Create or review technical design documentation to ensure the accurate development of reporting solutions.

    AI DOES

    Drafts templates, flags missing sections, and checks documentation for gaps or inconsistencies

    YOU DO

    Verify that business rules, data sources, and stakeholder needs are correctly captured

  26. TASK 26

    Analyze technology trends to identify markets for future product development or to improve sales of existing products.

    AI DOES

    Scans industry reports, news, and data to surface emerging technology trends and market signals fast

    YOU DO

    Apply business context, customer knowledge, and strategic judgment to decide which trends are worth pursuing

  27. TASK 27

    Deliver oral or written presentations of the results of mathematical modeling and data analysis to management or other end users.

    AI DOES

    Drafts slides, writes summaries, and turns raw results into plain-language narratives

    YOU DO

    Read your audience and shape the message around what decisions they need to make

  28. TASK 28

    Monitor work productivity or quality to ensure compliance with standard operating procedures.

    AI DOES

    Tracks productivity metrics, flags deviations from SOPs, and generates compliance reports automatically

    YOU DO

    Judge root causes, coach team members, and own final compliance decisions

  29. TASK 29

    Develop project-specific data management plans that address areas such as coding, reporting, or transfer of data, database locks, and work flow processes.

    AI DOES

    Drafts template sections, suggests standard wording, and flags missing components

    YOU DO

    Apply trial-specific context, negotiate with stakeholders, and take regulatory responsibility

  30. TASK 30

    Evaluate processes and technologies, and suggest revisions to increase productivity and efficiency.

    AI DOES

    Researches current technologies, benchmarks processes, and drafts improvement recommendations

    YOU DO

    Weigh regulatory risk, team readiness, and real-world clinical data constraints before deciding

  31. TASK 31

    Develop technical specifications for data management programming and communicate needs to information technology staff.

    AI DOES

    Drafts technical specification documents, suggests standard data formats, and flags missing requirements

    YOU DO

    Confirm clinical accuracy, negotiate with IT staff, and sign off on final specs

  32. TASK 32

    Design and validate clinical databases, including designing or testing logic checks.

    AI DOES

    Drafts database schemas, suggests edit checks, and flags common logic errors

    YOU DO

    Validate that every rule matches the clinical protocol and meets regulatory standards

  33. TASK 33

    Train staff on technical procedures or software program usage.

    AI DOES

    Drafts training guides, quizzes, slide decks, and step-by-step tutorials for clinical software

    YOU DO

    Read learner confusion, answer unexpected questions, and adapt training in real time

  34. TASK 34

    Provide support and information to functional areas such as marketing, clinical monitoring, and medical affairs.

    AI DOES

    Drafts data summaries, answers common questions, and formats reports for different teams

    YOU DO

    Judge what information each team truly needs and build trust across departments

  35. TASK 35

    Confer with end users to define or implement clinical system requirements such as data release formats, delivery schedules, and testing protocols.

    AI DOES

    Drafts requirement templates, summarizes meeting notes, and suggests standard data formats or testing checklists

    YOU DO

    Lead the conversations, read stakeholder needs, and make final calls on what will actually work in practice

  36. TASK 36SOURCE-BACKED TASK

    Prepare appropriate formatting to data sets as requested.

    AI DOES

    Converts data between formats and applies templates

    YOU DO

    Choose the right structure for the purpose

    About this analysis

    59% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  37. TASK 37

    Supervise the work of data management project staff.

    AI DOES

    Tracks task progress, surfaces bottlenecks, and drafts performance notes

    YOU DO

    Build trust, read team dynamics, and make judgment calls about people

  38. TASK 38SOURCE-BACKED TASK

    Generate standard or custom reports summarizing business, financial, or economic data for review by executives, managers, clients, and other stakeholders.

    AI DOES

    Creates charts, tables, and draft summaries from raw data

    YOU DO

    Shape the story and key insights

    About this analysis

    49% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  39. TASK 39SOURCE-BACKED TASK

    Provide technical support for existing reports, dashboards, or other tools.

    AI DOES

    Diagnoses common issues and suggests standard fixes

    YOU DO

    Understand the real business problem

    About this analysis

    49% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  40. TASK 40SOURCE-BACKED TASK

    Process clinical data, including receipt, entry, verification, or filing of information.

    AI DOES

    Enters and organizes patient data automatically

    YOU DO

    Verify accuracy and flag critical issues

    About this analysis

    47% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  41. TASK 41SOURCE-BACKED TASK

    Collect business intelligence data from available industry reports, public information, field reports, or purchased sources.

    AI DOES

    Scans reports and extracts key data points

    YOU DO

    Decide what insights drive strategy

    About this analysis

    47% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  42. TASK 42SOURCE-BACKED TASK

    Create business intelligence tools or systems, including design of related databases, spreadsheets, or outputs.

    AI DOES

    Generates dashboards, queries, and database schemas

    YOU DO

    Define what insights the business needs

    About this analysis

    45% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  43. TASK 43SOURCE-BACKED TASK

    Prepare data analysis listings and activity, performance, or progress reports.

    AI DOES

    Generates charts, tables, and draft reports from data

    YOU DO

    Interpret findings and tell the story

    About this analysis

    41% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  44. TASK 44SOURCE-BACKED TASK

    Synthesize current business intelligence or trend data to support recommendations for action.

    AI DOES

    Analyzes patterns and generates trend summaries

    YOU DO

    Connect insights to real business decisions

    About this analysis

    40% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  45. TASK 45SOURCE-BACKED TASK

    Analyze clinical data using appropriate statistical tools.

    AI DOES

    Runs statistical tests and generates initial analysis

    YOU DO

    Choose the right methods and interpret results

    About this analysis

    39% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  46. TASK 46SOURCE-BACKED TASK

    Identify and analyze industry or geographic trends with business strategy implications.

    AI DOES

    Processes data and spots initial patterns

    YOU DO

    Judge which trends matter for strategy

    About this analysis

    37% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  47. TASK 47SOURCE-BACKED TASK

    Contribute to the compilation, organization, and production of protocols, clinical study reports, regulatory submissions, or other controlled documentation.

    AI DOES

    Drafts sections and formats documents to standards

    YOU DO

    Ensure compliance and scientific accuracy

    About this analysis

    36% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  48. TASK 48SOURCE-BACKED TASK

    Maintain library of model documents, templates, or other reusable knowledge assets.

    AI DOES

    Organizes, tags, and suggests relevant templates

    YOU DO

    Curate quality and decide what stays

    About this analysis

    34% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  49. TASK 49SOURCE-BACKED TASK

    Develop or select specific software programs for various research scenarios.

    AI DOES

    Suggests tools based on research requirements

    YOU DO

    Judge fit for your specific context

    About this analysis

    34% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  50. TASK 50SOURCE-BACKED TASK

    Analyze competitive market strategies through analysis of related product, market, or share trends.

    AI DOES

    Processes data and identifies market trends quickly

    YOU DO

    Interpret what trends mean for strategy

    About this analysis

    31% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  51. TASK 51SOURCE-BACKED TASK

    Design forms for receiving, processing, or tracking data.

    AI DOES

    Creates form layouts and suggests field types

    YOU DO

    Design flows that users actually want

    About this analysis

    30% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  52. TASK 52SOURCE-BACKED TASK

    Disseminate information regarding tools, reports, or metadata enhancements.

    AI DOES

    Drafts announcements and formats updates quickly

    YOU DO

    Choose timing, audience, and key messages

    About this analysis

    9% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  53. TASK 53SOURCE-BACKED TASK

    Read technical literature and participate in continuing education or professional associations to maintain awareness of current database technology and best practices.

    AI DOES

    Summarizes papers and highlights key trends

    YOU DO

    Judge what applies to your work

    About this analysis

    8% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

  54. TASK 54SOURCE-BACKED TASK

    Communicate with customers, competitors, suppliers, professional organizations, or others to stay abreast of industry or business trends.

    AI DOES

    Monitors news and summarizes industry reports

    YOU DO

    Build relationships and ask smart questions

    About this analysis

    0% · Methodology: exposure-v1 · Analysis date: unavailable · Confidence: medium

START LEARNING AI

Practice AI.
Use your judgment.

There are 54 short lessons for this job, about 15 minutes each. Each one starts from a task above and teaches the part that stays with you.

Try a free lesson
Data sources and attribution

This page includes information from the O*NET 30.3 Database by the U.S. Department of Labor, Employment and Training Administration (USDOL/ETA). Used under the CC BY 4.0 license. rtwoai has modified all or some of this information. USDOL/ETA has not approved, endorsed, or tested these modifications. O*NET® is a trademark of USDOL/ETA.

Wage and employment figures are from the U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (2025). Job outlook is from the BLS Employment Projections program, 2024–34. BLS data are in the public domain. BLS has not reviewed or endorsed this page.

Task collaboration data is from the Anthropic Economic Index, release 2025-03-27, used under the CC BY 4.0 license. Anthropic has not reviewed or endorsed this page.

How we describe AI use in each task is our own characterization, not a measurement of whether AI can replace this job. Microsoft note that their metrics "should not be misconstrued or misrepresented as measuring the ability of AI to replace jobs," and that there is more to an occupation than the sum of its tasks.