TGThe Task GraphRole intelligence
Browse rolesRequest a roleHow it works

The Task Graph

Understand how AI is changing work — built from public government databases with transparent methods and plain-language explanations.

Browse rolesHow it worksRequest a rolePrivacy
Browse roles/Customer Service Representatives
Role-specific view13 role tasks

Role summary

Customer Service Representatives

AI changes this role through complaint triage, records, and routine updates, while difficult resolutions and customer trust stay human-led.

Interact with customers to provide basic or scripted information in response to routine inquiries about products and services. May handle and resolve general complaints. Excludes individuals whose duties are primarily installation, sales, repair, and technical support.

Explore tasksRequest this role or another
Account AdjusterAccount RepresentativeAdjustment ClerkAutomotive Service AdvisorAutomotive Service WriterBill AdjusterCall Center RepresentativeClaims Customer Service Representative (Claims CSR)

Work impacted by AI

Moderate AI support30–45%

AI can help with a meaningful share of the work, while people still own judgment, review, and relationships.

4 tasks most likely to change

AI may reach parts of the task, not erase the task.

13 tasks have external evidence

Task-level matches are marked in the details.

Not a job-loss prediction or a claim that this work should be fully automated.

Source coverage and task evidence are shown in the task section and evidence section below.

How this estimate is calculated

The estimate averages the analyzed task-level AI impact signals for this role. Each task starts with stored task analysis, then uses direct public task evidence when an exact task match is available.

13 tasks are included; 13 have matched external task evidence. Role-level benchmarks are shown below as context, not pushed into individual task estimates.

Tasks in this role

Which tasks may change with AI support

Scan the work areas, choose a task, then inspect what changes, what stays human-led, and which source signals are attached.

List and detail viewExternal evidence

Reviewed task path

Start with the reviewed examples, then explore every task in the role.

Find tasks

13 of 13 tasks

Search directly, or filter for the tasks most worth inspecting first. Work area is a secondary filter.

2 strong AI-support tasks13 tasks with direct evidence

Selected task

Obtain and examine all relevant information to assess validity of complaints and to determine possible causes, such as extreme weather conditions that could increase utility bills.

Emerging AI work-change signal
Workflow redesignAI impact 50%1 source signal

Why this task is shown

i

Select a node; details update below.

Selected task: Obtain and examine all relevant information to assess validity of complaints and to determine possible caus.... Click a node to inspect the source path for this task.

8 path nodes1 evidence

Task

The selected work activity

Work concept

Plain-language bridge

Estimate

AI work-change read

Evidence

Matched public support

Source

Where signals came from

Evidence path

Select a node to inspect the evidence path.

Selected node

This section updates when you choose a node in the path.

Task

Obtain and examine all relevant information to assess validity of complaints and to determine possible causes, such as extreme weather conditions that could increase utility bills.

The work looks too variable for reliable end-to-end automation. It combines structured inputs with case judgment. Formal accountability pressure looks limited. People still need to stay close to the workflow even if parts are automated. This task blends analytical reasoning and decision making work. This task mixes technical limits with delivery constraints, so the best first move is usually narrow AI assistance rather than full automation.

Why linked

maps to

AI work-change
50%
Workflow automation
29%

Connected path

  1. Obtain and examine all relevant information to assess validity of compl...
  2. Analytical Reasoning
  3. Decision Making
  4. AI work-change estimate
  5. Workflow automation estimate
  6. OpenAI GPTs are GPTs
  7. OpenAI GPTs are GPTs
  8. The Task Graph analysis

Supporting details

2

Where AI reaches

1 matched public source signal is visible for this task. The signal suggests AI may reach preparation, review, or parts of the workflow, but the whole task still needs context.

What changes

The work looks too variable for reliable end-to-end automation.

What stays human

Automation needs narrow scope and visible review because delivery constraints still matter.

Next move

Start with workflow redesign

Look for handoffs, approvals, or tool changes before trying to automate the task itself.

Sources

Evidence attached to this task

1 matched

OpenAI GPTs are GPTs

Exact O*NET task-id match

medium

How this source is linked

This source row is matched to the same O*NET occupation task used in The Task Graph. It is an exact task-id match, not a fuzzy text guess.

Confidence
95%
What the value means
Model exposure signal; it does not mean full task automation.

OpenAI records a medium public AI source signal for this task (50 / 100). This is best read as model exposure and possible AI assistance, not as a guarantee of full automation.

View source

Why this signal is shown

1 external source signal matched to this task. The visible signal combines stored task analysis with matched public task evidence when a direct match exists.

Evidence trail

Work concepts

Analytical ReasoningDecision Making

Evidence records

OpenAI GPTs are GPTs

OpenAI records a medium public AI source signal for this task (50 / 100). This is best read as model exposure and possible AI assistance, not as a guarantee of full automation.

medium

Sources

OpenAI GPTs are GPTsThe Task Graph analysis

Evidence

What the page is grounded in

The task view keeps task-level evidence close to each task. This section keeps broader role context and source records available when you want to inspect them.

Role-level benchmarks

These records describe the occupation as a whole. They provide context for the top estimate, but task-level evidence is only used on a task when it directly matches that task.

AI USAGE

ANTHROPIC

Mapped directly to this occupation.

high

AI USAGE

MICROSOFT

Mapped directly to this occupation.

medium

AI EXPOSURE

OPENAI

Mapped directly to this occupation.

high

Supporting role signals

AI EXPOSURE

INTERNAL · INTERNAL MODEL

low

ROUTINE INTENSITY

OECD · GROUP MAPPING

medium

HUMAN INTERACTION LEVEL

INTERNAL · INTERNAL MODEL

medium

TASK COMPLEXITY

INTERNAL · INTERNAL MODEL

high

Occupation source records

These are the occupation records used for the role structure. Public AI evidence links are shown beside the matched task evidence above.

ONET

O*NET

View source record

Important to know

Task-level evidence can shape a task estimate only when it is matched directly to that task. Role-level benchmarks stay as context because they describe the occupation as a whole.

Task sources

Tasks

O*NET · O*NET Task Statements.txt · db_30_0_text

Need a different view?

Request a role analysis

Ask for this role, another role, or a version adapted to your actual tasks. No card is charged from this page.

Request analysis