Computer Network Support Specialists

15-1231.00
Median wage $76,220/yr146,190 employed (US)Rank #58 of 923 scored · top 6% by substitution

Analyze, test, troubleshoot, and evaluate existing network systems, such as local area networks (LAN), wide area networks (WAN), cloud networks, servers, and other data communications networks. Perform network maintenance to ensure networks operate correctly with minimal interruption.

Sub-scores

0–100 · band = confidence interval from rater disagreement

Substitution56
Exposure52
Augmentation76

Substitution — the headline: capability discounted by cost, barriers and adoption.

Exposure — technical capability alone, regardless of whether anyone deploys it.

Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.

Tasks on the substitution scale

26 rated tasks, binned by substitution score.

Position among all scored occupations

Distribution of 923 occupation scores; the marker is this occupation.

Tasks with substitution ≥ 70

27%

Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.

Why this score

The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.

Task automatabilityw 35%51

panel mean rating 3.1/5 → substitution pressure 51/100

Technical feasibility todayw 20%52

panel mean rating 3.1/5 → substitution pressure 52/100

Cost vs. human wagew 15%57

panel mean rating 3.3/5 → substitution pressure 57/100

Adoption barriersw 20%inverted — strong barriers lower the score66

panel mean rating 2.4/5 (barrier strength) → substitution pressure 66/100

Sector adoption velocityw 10%59

panel mean rating 3.4/5 → substitution pressure 59/100

Task breakdown (26 tasks)

Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.

Back up network data.

93

CI 86–100 · exposure 92 · augmentation 75 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5Automated network backup has been standard practice in enterprise IT and digitized organizations for decades; adoption is nearly universal in medium-to-large firms and cloud environments, with continuous expansion into SMBs and hybrid infrastructure.
Sector adoption velocityclaude-sonnet-55/5IT infrastructure and network administration are among the most digitized functions with automated backup and monitoring tools already deeply embedded in production environments.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-driven backup systems significantly assist network specialists by automating scheduling, detecting anomalies, generating compliance reports, and flagging failed backups; the specialist role shifts to monitoring, troubleshooting, and policy adjustment rather than manual execution, substantially raising productivity.
Augmentation potentialclaude-sonnet-54/5Even where full automation exists, AI-driven monitoring and anomaly detection assist specialists in verifying backup integrity, troubleshooting failures, and optimizing schedules, though a human remains involved in oversight and incident response.
Task automatabilityclaude-haiku-4-5-202510015/5Network data backup is highly structured and rule-based, requiring execution of predefined procedures (scheduling, incremental backups, verification). Current AI and automation tools (enterprise backup software with agent-based orchestration) can perform this end-to-end with >50% time savings through scheduling, monitoring, and validation with minimal human intervention.
Task automatabilityclaude-sonnet-54/5Backing up network data is a well-defined, repetitive process that can be fully scripted and orchestrated with automated backup software and scheduling systems, requiring little ongoing human judgment.
Adoption barriersclaude-haiku-4-5-202510012/5While backup execution can be fully automated, some regulatory frameworks (HIPAA, SOX, GDPR) may require human attestation or compliance sign-off, and organizational policy often mandates human oversight of backup validation. However, no hard legal requirement exists that a human must manually execute the backup task itself.
Adoption barriersclaude-sonnet-51/5No licensing, regulatory, or human-contact requirements block automated backups; organizations widely already use automated tools without special authorization.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated backup systems operate at marginal cost per GB/backup cycle after initial setup, whereas human network specialists command $50k–$80k+ annually; AI/automation cost per equivalent backup operation is orders of magnitude cheaper, especially at organizational scale.
Cost vs. human wageclaude-sonnet-55/5Automated backup software costs a small fraction of a technician's time to manually initiate and monitor backups, and once configured runs unattended at near-zero marginal cost.
Technical feasibility todayclaude-haiku-4-5-202510015/5Mature, production-grade backup solutions (Veeam, Commvault, NetBackup, Arcserve, cloud-native backup services) demonstrably perform automated network data backup at scale across thousands of organizations globally with reliable execution, scheduling, and reporting.
Technical feasibility todayclaude-sonnet-55/5Mature enterprise backup products (e.g., automated snapshot, cloud backup, disaster recovery orchestration tools) reliably perform scheduled backups in production at scale across industries today.

Maintain logs of network activity.

91

CI 81–100 · exposure 87 · augmentation 63 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5Network logging automation is standard practice across information technology and cloud-native sectors; enterprise adoption of centralized logging platforms is near-universal, representing deep, fast displacement of manual log management.
Sector adoption velocityclaude-sonnet-55/5IT and network operations is a highly digitized sector where automated logging and monitoring tools have been standard practice for years, representing mature, widespread adoption.
Augmentation potentialclaude-haiku-4-5-202510013/5While humans are largely removed from routine log maintenance, AI-assisted log analysis (anomaly detection, pattern recognition, automated alerting) meaningfully enhances analysts' ability to investigate and respond to network events, providing useful assistance within a mostly automated workflow.
Augmentation potentialclaude-sonnet-54/5AI-enhanced log analysis tools help specialists quickly parse, summarize, and flag anomalies in large log volumes, significantly boosting their efficiency even when humans still review and act on findings.
Task automatabilityclaude-haiku-4-5-202510015/5Maintaining network activity logs is fully automatable today; modern network monitoring and logging systems (ELK stack, Splunk, cloud-native solutions) continuously collect, parse, and store logs without human intervention, easily meeting the ≥50% time-saving threshold.
Task automatabilityclaude-sonnet-54/5Log maintenance is largely mechanical—collecting, aggregating, and structuring network activity data—which existing monitoring tools and scripts already automate almost entirely, with AI adding parsing and anomaly summarization on top.
Adoption barriersclaude-haiku-4-5-202510012/5While regulatory compliance (HIPAA, SOC 2, PCI-DSS) requires audit trails and proper log retention, these requirements are typically satisfied by automated systems rather than mandating human involvement; oversight and configuration remain necessary but do not prevent automation.
Adoption barriersclaude-sonnet-51/5There is no licensing or liability requirement mandating a human perform routine log maintenance; it's a standard IT operations function with no regulatory human-sign-off requirement.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated logging infrastructure costs a fraction of the labor required for manual log maintenance; a single monitoring system handles logs across thousands of systems, making per-log-entry cost orders of magnitude cheaper than human effort.
Cost vs. human wageclaude-sonnet-55/5Automated logging agents and centralized log platforms cost far less per unit of data processed than manual log-keeping by a human specialist, especially at scale.
Technical feasibility todayclaude-haiku-4-5-202510015/5Mature, production-scale products for automated log collection and maintenance are widely deployed across enterprises; systems like Splunk, DataDog, and cloud providers' native logging services reliably perform this task at scale with minimal manual effort.
Technical feasibility todayclaude-sonnet-54/5Mature SIEM, network monitoring, and log management products (e.g., Splunk, SolarWinds, ELK stack) reliably automate log collection and retention in production environments today.

Document help desk requests and resolutions.

90

CI 80–100 · exposure 87 · augmentation 88 · importance 3.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5IT and help desk operations are highly digitized sectors with rapid AI adoption; many organizations already use AI for ticket triage, auto-categorization, and documentation at scale.
Sector adoption velocityclaude-sonnet-54/5IT service management and help desk software vendors have rapidly integrated AI summarization and auto-documentation features, reflecting fast adoption in this tech-forward sector.
Augmentation potentialclaude-haiku-4-5-202510014/5AI assists help desk staff by auto-populating fields, suggesting summaries, and generating resolution documentation in real-time, meaningfully raising their throughput and reducing manual typing without removing human review.
Augmentation potentialclaude-sonnet-55/5AI substantially speeds up documentation by auto-drafting summaries from interaction logs, letting technicians review and edit rather than write from scratch, a clear productivity boost while keeping human oversight.
Task automatabilityclaude-haiku-4-5-202510015/5Documenting help desk requests and resolutions is highly structured text capture and logging. Current AI systems can extract key information from tickets, emails, or chat transcripts, categorize issues, summarize resolutions, and populate documentation systems with >50% time savings and equal or better quality than manual entry.
Task automatabilityclaude-sonnet-54/5Documenting help desk requests and resolutions is largely a structured writing task—summarizing an interaction into a ticket—which current LLMs can do well, especially when integrated with ticketing systems that capture chat/call transcripts.
Adoption barriersclaude-haiku-4-5-202510011/5Documentation is a support function with no licensing requirements, liability barriers, or legal sign-off mandates. Organizations face minimal friction adopting automated logging and summarization.
Adoption barriersclaude-sonnet-51/5There is no licensing, legal, or safety requirement mandating a human write ticket documentation; it's an internal administrative task with minimal regulatory friction.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI inference costs for processing and documenting a ticket are typically cents per ticket, while human time to document (even briefly) costs dollars; AI is roughly one to two orders of magnitude cheaper all-in.
Cost vs. human wageclaude-sonnet-54/5Automated summarization and documentation via AI is cheap at scale compared to a technician spending manual time writing detailed notes for every ticket.
Technical feasibility todayclaude-haiku-4-5-202510015/5Deployed products like ServiceNow with AI modules, Zendesk's AI, and general LLM-based ticket summarization and classification systems are already performing this task reliably in production environments across thousands of organizations.
Technical feasibility todayclaude-sonnet-54/5Many ITSM platforms (ServiceNow, Zendesk, Freshdesk) now offer AI-generated ticket summaries and auto-categorization in production, though human review/editing is still common for accuracy.

Run monthly network reports.

87

CI 75–100 · exposure 87 · augmentation 88 · importance 3.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5IT and network operations are highly digitized sectors with fast, deep adoption of monitoring and automation platforms; automated reporting is now standard practice across enterprise and mid-market organizations.
Sector adoption velocityclaude-sonnet-54/5IT and network operations functions have widely adopted automated monitoring and reporting tools for years, representing a mature, fast-adopting use case within tech-forward sectors.
Augmentation potentialclaude-haiku-4-5-202510015/5Beyond full automation, AI-powered monitoring and analytics tools substantially augment human network specialists by surfacing anomalies, suggesting root causes, and automating alert aggregation, raising overall productivity.
Augmentation potentialclaude-sonnet-54/5AI tools can significantly speed up report generation, anomaly highlighting, and narrative summaries, letting specialists focus on analysis and remediation rather than manual compilation.
Task automatabilityclaude-haiku-4-5-202510015/5Running monthly network reports is a highly structured, repetitive task involving data collection from systems, aggregation, and formatting. Current AI agents and scheduled automation can fully execute this end-to-end—pulling logs, generating statistics, and formatting reports—delivering ≥50% time savings at equal quality.
Task automatabilityclaude-sonnet-54/5Generating monthly network reports is largely a data-pulling, formatting, and summarization task that can be scripted or automated using existing network management tools and AI-assisted reporting pipelines, meeting the time-saving threshold with modest setup.},
Adoption barriersclaude-haiku-4-5-202510012/5No licensing requirement mandates human sign-off on network reports; some organizations may prefer human review for accuracy or liability reasons, but technical barriers to full automation are minimal.
Adoption barriersclaude-sonnet-51/5There are no licensing, legal, or human-contact requirements around generating routine network reports; it's a purely administrative/technical task.
Cost vs. human wageclaude-haiku-4-5-202510015/5Once configured, automated reporting has near-zero marginal cost per month, while a human support specialist's loaded hourly wage for an hour of report generation is substantial, making AI orders of magnitude cheaper.
Cost vs. human wageclaude-sonnet-54/5Automated reporting tools and scripts run at a fraction of the cost of a technician's time once configured, though initial setup and occasional oversight add some cost.
Technical feasibility todayclaude-haiku-4-5-202510015/5Mature products (monitoring platforms like Splunk, Prometheus, Grafana, and cloud-native tools) reliably perform automated report generation in production at scale across thousands of organizations today.
Technical feasibility todayclaude-sonnet-54/5Mature network monitoring platforms (SolarWinds, PRTG, Nagios, etc.) already generate automated scheduled reports in production, and AI/LLM tools can further summarize and format these reliably today.

Monitor industry Web sites or publications for information about patches, releases, viruses, or potential problem identification.

82

CI 75–89 · exposure 80 · augmentation 100 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5IT security and operations teams have rapidly adopted automated threat monitoring, patch tracking, and vulnerability management systems; this is mainstream practice in enterprise and mid-market IT sectors.
Sector adoption velocityclaude-sonnet-54/5IT and cybersecurity functions have rapidly adopted automated threat intelligence and patch monitoring tools as standard practice in most organizations with network infrastructure.
Augmentation potentialclaude-haiku-4-5-202510015/5AI monitoring systems substantially augment human network specialists by pre-filtering threats, ranking severity, and alerting on emerging issues in real time, allowing humans to focus on response rather than discovery.
Augmentation potentialclaude-sonnet-55/5AI significantly enhances a specialist's ability to stay current by filtering, summarizing, and prioritizing vast amounts of security and patch information, letting humans focus on decision-making and remediation.
Task automatabilityclaude-haiku-4-5-202510014/5AI can automatically monitor web sites, parse news feeds, and identify relevant patches, releases, and viruses with high reliability today. However, the task typically requires some human judgment to prioritize findings and filter false positives, preventing a full 5 rating.
Task automatabilityclaude-sonnet-54/5AI-based monitoring tools and agents can continuously scan vendor sites, CVE feeds, and security bulletins, aggregating and summarizing relevant information with substantial time savings over manual browsing.
Adoption barriersclaude-haiku-4-5-202510012/5Few hard legal or regulatory barriers exist to automating this monitoring task. Organizations may prefer human review for validation, but nothing prevents substitution of the core monitoring activity.
Adoption barriersclaude-sonnet-51/5No licensing or legal requirement mandates a human to perform this monitoring; it's a routine informational task with no liability barrier to automation.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated monitoring systems cost a small fraction of a human specialist's loaded wage ($80k+), especially when amortized across enterprise deployments, making AI orders of magnitude cheaper.
Cost vs. human wageclaude-sonnet-54/5Automated feed aggregation and AI summarization tools cost a small fraction of a technician's hourly wage for continuous monitoring, though subscription costs for premium threat intel feeds add some expense.
Technical feasibility todayclaude-haiku-4-5-202510015/5Deployed products (security monitoring tools, threat intelligence platforms, automated RSS/news aggregators with AI filtering) reliably perform this task at scale in production IT environments across many organizations.
Technical feasibility todayclaude-sonnet-54/5Deployed products (threat intelligence platforms, RSS/AI summarization tools, vulnerability management systems like Tenable, Recorded Future) already perform this monitoring reliably in production, though final triage often still involves human review.

Document network support activities.

73

CI 67–79 · exposure 70 · augmentation 100 · importance 4.1/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5IT and support operations are highly digitized sectors with rapid AI adoption; major enterprises already deploy automated ticket logging, documentation generation, and incident summarization at scale. This is a mainstream practice in large organizations and growing in mid-market.
Sector adoption velocityclaude-sonnet-53/5IT/network support functions are moderately digitized with growing use of AI copilots for ticket summarization and documentation, but many organizations still rely on manual entry, especially in smaller firms.
Augmentation potentialclaude-haiku-4-5-202510015/5AI substantially augments support specialists by auto-generating drafts, suggesting relevant prior incidents, and formatting data, allowing humans to focus on analysis and exception handling. The human remains in the loop while AI dramatically accelerates documentation tasks.
Augmentation potentialclaude-sonnet-55/5AI drafting and summarization tools substantially speed up documentation tasks, letting technicians review and edit rather than write from scratch, a clear productivity boost while the human remains in the loop.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI can automatically generate structured documentation of support activities by parsing logs, tickets, and chat transcripts, reducing manual writing and categorization by 60–80%. Most of the task involves converting existing data into documented form, which modern LLMs and RPA tools handle reliably, though some human review may still be needed for accuracy and completeness.
Task automatabilityclaude-sonnet-54/5Documenting network support activities (tickets, changes, resolutions) is largely text generation and summarization from structured/unstructured inputs, which LLMs handle well with modest integration into ticketing systems.“ Full automation requires connecting to logs and ticketing tools, but the core task meets the time-saving bar.
Adoption barriersclaude-haiku-4-5-202510012/5Documentation is a support task with low legal/regulatory barriers to automation and no licensing requirement. Light oversight may be preferred for accuracy and compliance, but nothing legally mandates human authorship, so adoption friction is minimal.
Adoption barriersclaude-sonnet-52/5No licensing or legal requirement mandates human authorship of internal documentation, though organizational policies may require technician sign-off for accuracy and accountability.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI-driven documentation automation costs pennies per incident via LLM inference and logging infrastructure, compared to specialist labor at $50–80/hour spent manually writing reports. The cost differential is at least an order of magnitude in favor of AI.
Cost vs. human wageclaude-sonnet-54/5Generating documentation via AI is far cheaper per instance than a technician manually writing detailed notes, though some oversight cost remains for accuracy verification.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed products like ServiceNow, Jira, and AI-assisted documentation tools already perform automated logging and summary generation in production environments. AI can extract ticket data and generate incident reports at scale with good reliability, though some edge cases and complex narrative summaries may require human refinement.
Technical feasibility todayclaude-sonnet-53/5AI-assisted documentation tools and copilots integrated into ITSM platforms (e.g., ServiceNow, Freshservice) exist and are used in production, but coverage of technical nuance and accuracy still requires human review, limiting full reliability.

Identify the causes of networking problems, using diagnostic testing software and equipment.

72

CI 55–90 · exposure 70 · augmentation 75 · importance 4.2/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5Large enterprises and cloud platforms have rapidly adopted AI-driven network monitoring and anomaly detection over the past 5–10 years. Production deployment is standard practice in information and telecom sectors, indicating fast, deep adoption.
Sector adoption velocityclaude-sonnet-53/5IT operations sectors are moderately fast adopters of AIOps tools, with growing production use, but many organizations still rely primarily on human network engineers for diagnosis.
Augmentation potentialclaude-haiku-4-5-202510014/5AI diagnostic tools significantly augment human specialists by automating tedious testing cycles, surfacing correlations across logs, and narrowing problem scope. Specialists remain in the loop to interpret ambiguous findings and implement fixes, but AI productivity gains are substantial.
Augmentation potentialclaude-sonnet-54/5Diagnostic software with AI-driven pattern recognition significantly speeds up root-cause identification and reduces mean-time-to-resolution, making it a strong productivity multiplier for specialists who remain in the loop.
Task automatabilityclaude-haiku-4-5-202510015/5Network diagnostics are highly structured, data-driven tasks where AI can run automated testing (ping, traceroute, packet analysis), correlate logs, and identify root causes with deterministic logic. Current systems can often match or exceed human speed and accuracy at fault isolation without human intervention.
Task automatabilityclaude-sonnet-53/5AI can analyze logs, run diagnostics, and suggest likely root causes for common networking issues, but complex or novel infrastructure problems still require human investigation and physical access.5 Full end-to-end diagnosis across heterogeneous environments isn't yet reliably automatable.
Adoption barriersclaude-haiku-4-5-202510012/5No legal licensing requirement exists for automated network diagnostics; organizations routinely deploy AI monitoring without specialist sign-off. Main friction is organizational preference to pair automated diagnosis with human validation for critical systems, but no hard barrier prevents substitution.
Adoption barriersclaude-sonnet-52/5No licensing requirement mandates a human specifically, though critical infrastructure changes may require sign-off from certified staff; organizational risk aversion around network outages creates moderate friction.
Cost vs. human wageclaude-haiku-4-5-202510015/5AI-driven network monitoring and diagnostic agents run continuously at marginal inference cost after initial setup, orders of magnitude cheaper than employing full-time human specialists to perform the same detection and isolation tasks.
Cost vs. human wageclaude-sonnet-53/5AI-assisted diagnostic tools reduce time spent on triage but still require licensing costs and human oversight, making the net cost comparable to a skilled specialist rather than dramatically cheaper.
Technical feasibility todayclaude-haiku-4-5-202510014/5Mature monitoring and diagnostic tools (Cisco, Splunk, NetBrain, Elastic) with AI/ML components already perform automated anomaly detection and root-cause analysis in production. Some systems require human confirmation on ambiguous cases, but diagnosis itself is demonstrably deployed at scale.
Technical feasibility todayclaude-sonnet-53/5AIOps and network monitoring products (e.g., Cisco AI Network Analytics, Splunk, ThousandEyes) deploy anomaly detection and root-cause suggestions in production, but they often require human validation and struggle with edge cases or multi-vendor complexity.

Research hardware or software products to meet technical networking or security needs.

69

CI 59–80 · exposure 62 · augmentation 100 · importance 3.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5IT and networking sectors are digitally mature and rapidly deploying AI-assisted tools for resource evaluation and decision support. Product research automation fits established patterns in cloud infrastructure and SaaS evaluation workflows.
Sector adoption velocityclaude-sonnet-53/5IT and network support functions are moderately digitized and increasingly using AI search/summarization tools, but broad organizational adoption for technical procurement research is still uneven.
Augmentation potentialclaude-haiku-4-5-202510015/5AI transforms productivity on this task by instantly aggregating vendor specs, compliance matrices, benchmark data, and cost comparisons while the human engineer applies domain judgment to final selection. This is a high-productivity augmentation scenario.
Augmentation potentialclaude-sonnet-55/5AI substantially speeds up gathering and synthesizing information on hardware/software options, letting specialists focus on final evaluation and decision-making.
Task automatabilityclaude-haiku-4-5-202510014/5AI can systematically research and compare hardware/software products against specified networking or security requirements, compile technical specifications, and generate vendor recommendation reports with substantial time savings. However, the task may require some human validation for edge cases, emerging products, or organization-specific constraints, preventing a full 5 rating.
Task automatabilityclaude-sonnet-53/5AI can gather, summarize, and compare product specs, reviews, and compatibility data quickly, but final vetting against specific network architecture and security requirements still needs human judgment and validation., limiting full end-to-end automation.
Adoption barriersclaude-haiku-4-5-202510012/5Few regulatory or legal barriers prevent automated product research; however, organizational preference for human judgment on critical infrastructure decisions and vendor relationships creates modest friction against full automation.
Adoption barriersclaude-sonnet-51/5No licensing or regulatory requirement mandates a human perform product research; it's an informational task with low liability exposure at this stage.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-driven product research has very low marginal cost per research task (pennies per query) compared to labor costs for a specialist's research time, representing roughly a 10-50x cost advantage depending on search depth and integration overhead.
Cost vs. human wageclaude-sonnet-54/5AI-assisted research (LLM queries, comparison summaries) is far cheaper than hours of manual specification review, though human validation adds some cost back.
Technical feasibility todayclaude-haiku-4-5-202510014/5Deployed systems (LLMs with web search, specialized product databases, vendor APIs) reliably perform comparative research on networking and security products in production environments. Performance is strong on well-documented products but may have gaps on niche or newly released solutions.
Technical feasibility todayclaude-sonnet-53/5AI chat and search tools (e.g., with web browsing or RAG) are commonly used to research products today, but they can produce outdated or inaccurate specs, so specialists still verify claims against vendor documentation.

Analyze network data to determine network usage, disk space availability, or server function.

69

CI 55–82 · exposure 67 · augmentation 88 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5IT and cloud-native organizations have rapidly adopted automated monitoring, alerting, and AI-driven analytics over the past decade; this is mainstream practice in high-digitization sectors like software, finance, and cloud services.
Sector adoption velocityclaude-sonnet-53/5IT operations and DevOps are moderately fast adopters of AIOps and monitoring analytics tools, but many organizations still rely on manual dashboards and ad hoc analysis rather than fully automated pipelines.
Augmentation potentialclaude-haiku-4-5-202510015/5AI-powered dashboards, anomaly detection, and natural-language summaries of network data significantly enhance specialist productivity by filtering noise, flagging issues, and automating routine reporting while humans focus on interpretation and remediation.
Augmentation potentialclaude-sonnet-54/5AI-powered dashboards, anomaly detection, and predictive analytics substantially speed up a specialist's ability to interpret network usage and capacity data while the human still makes final judgments.
Task automatabilityclaude-haiku-4-5-202510014/5Current AI systems can readily analyze network logs, disk usage metrics, and server performance data to identify patterns, bottlenecks, and anomalies. Automated dashboards and alerting systems already perform much of this analysis, and LLMs/agents can interpret raw data and generate summaries, likely saving >50% of time on routine analysis tasks.
Task automatabilityclaude-sonnet-53/5AI can analyze network logs, usage metrics, and disk/server telemetry to surface patterns and anomalies, but determining actionable conclusions and root causes often requires human judgment and system-specific context. With proper monitoring tooling and setup, roughly half the analytical workload could be offloaded to AI.
Adoption barriersclaude-haiku-4-5-202510012/5While many organizations require human review and sign-off on infrastructure changes, network analysis itself is not legally restricted and automation is already pervasive; minimal regulatory or licensing barriers exist for this task specifically.
Adoption barriersclaude-sonnet-52/5No licensing requirement blocks AI-assisted analysis, though organizational risk tolerance around infrastructure decisions and need for accountable IT staff creates some friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510014/5Cloud-based monitoring platforms operate at scale with per-agent or subscription costs often lower than a full-time specialist salary, especially when amortized across multiple servers and sites; overhead is minimal.
Cost vs. human wageclaude-sonnet-53/5AI-based monitoring tools reduce manual log review time but require licensing, integration, and ongoing tuning costs, plus human oversight, making the net savings moderate rather than order-of-magnitude cheaper.
Technical feasibility todayclaude-haiku-4-5-202510015/5Production-grade monitoring and analytics tools (Datadog, New Relic, Splunk, Prometheus) are widely deployed in enterprises and reliably perform this task at scale, including automated anomaly detection and performance reporting.
Technical feasibility todayclaude-sonnet-53/5Network monitoring products with AI-driven analytics (e.g., anomaly detection, capacity forecasting) are deployed in production, but they still require human interpretation and tuning, and coverage of edge cases or complex infrastructure varies.

Create or revise user instructions, procedures, or manuals.

66

CI 59–72 · exposure 58 · augmentation 88 · importance 3.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Technology and information-sector companies are piloting AI documentation tools, but adoption remains piecemeal; widespread production deployment is not yet standard practice in network support teams.
Sector adoption velocityclaude-sonnet-53/5IT and technical support functions are moderately fast adopters of AI writing tools, with common pilots and growing use of AI-assisted documentation, though full deployment for authoritative manuals is still uneven.
Augmentation potentialclaude-haiku-4-5-202510014/5AI is highly effective at assisting specialists by drafting templates, auto-generating procedure steps from network logs or configurations, and flagging inconsistencies, substantially accelerating manual creation and revision while the human retains quality control.
Augmentation potentialclaude-sonnet-55/5AI is highly effective at drafting, restructuring, and updating instructional content, letting the specialist focus on technical accuracy and final review, substantially boosting throughput.
Task automatabilityclaude-haiku-4-5-202510013/5AI can generate or revise procedural documentation drafts efficiently and could save 40–60% of time on routine manual updates, but requires human expertise to validate technical accuracy, organizational consistency, and appropriate detail levels for target audiences.
Task automatabilityclaude-sonnet-54/5Drafting and revising technical documentation from source material (config specs, ticket histories, SME input) is well within current LLM capabilities, often exceeding the 50% time-saving bar with light human review. The main remaining effort is verifying accuracy against actual systems.
Adoption barriersclaude-haiku-4-5-202510012/5No legal or licensing barrier prevents AI-assisted documentation; main friction is organizational preference for human sign-off and customer expectations around manual quality, but these are weak adoption barriers.
Adoption barriersclaude-sonnet-51/5There is no licensing or regulatory requirement that a human write internal network documentation, and no liability regime blocking AI-assisted authoring.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI inference for drafting and revision is inexpensive compared to loaded specialist wages, though integration and review overhead partially offset the savings; the ratio still favors automation for bulk documentation work.
Cost vs. human wageclaude-sonnet-54/5Generating and revising drafts via LLMs costs a small fraction of a specialist's hourly wage, even after factoring in review time, making AI substantially cheaper for the bulk of the writing task.
Technical feasibility todayclaude-haiku-4-5-202510013/5LLM-based documentation tools exist and are deployed in some organizations, but they frequently require substantial human review and editing to ensure correctness in technical contexts, making them reliable only with material oversight.
Technical feasibility todayclaude-sonnet-53/5AI writing assistants and documentation tools (e.g., Confluence AI, GitHub Copilot for docs) are used in production for drafting, but reliable, accurate technical manuals still require human verification, so full end-to-end reliability is not yet standard.

Create or update technical documentation for network installations or changes to existing installations.

63

CI 54–72 · exposure 58 · augmentation 88 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Technology and IT departments are early-to-mid adopters of AI tooling, with pilots and some production use in documentation automation, but adoption remains inconsistent across firm sizes and sectors. Large enterprises pilot solutions; mid-market and smaller firms lag due to integration cost and change management inertia.
Sector adoption velocityclaude-sonnet-53/5IT and technical documentation workflows are increasingly using AI drafting tools, but many network support teams still rely on manual documentation processes and pilots are more common than full production use.
Augmentation potentialclaude-haiku-4-5-202510014/5AI significantly assists specialists by drafting initial documentation from network schemas and change logs, auto-generating diagrams, and flagging inconsistencies, allowing humans to focus on review, refinement, and sign-off. This augmentation model is already common in practice and measurably increases documentation throughput and consistency.
Augmentation potentialclaude-sonnet-55/5AI substantially speeds up drafting, summarizing changes, and maintaining consistent documentation formats while the specialist verifies technical accuracy and completeness.
Task automatabilityclaude-haiku-4-5-202510013/5AI can draft and update technical documentation from network logs, change tickets, and specifications with significant time savings, but requires domain expertise review and organization-specific customization. Current systems handle ~50% of the work (initial drafts, formatting, basic coherence), but human specialists must verify accuracy, fill context gaps, and ensure compliance with internal standards.
Task automatabilityclaude-sonnet-54/5AI language models can draft, format, and update technical documentation from network configs, logs, or engineer notes with substantial time savings, though final review is often needed for accuracy.network.
Adoption barriersclaude-haiku-4-5-202510013/5Documentation often must be signed off by a responsible engineer or network manager due to liability and regulatory audit trails (SOX, HIPAA, PCI-DSS in regulated sectors), creating approval bottlenecks. However, no licensing requirement prevents automation of the drafting phase itself; friction comes from oversight and organizational sign-off workflows rather than legal prohibition.
Adoption barriersclaude-sonnet-51/5There is no licensing or legal requirement for a human to author network documentation, and no regulatory barrier prevents AI-assisted or AI-generated documentation.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI inference and integration costs are a small fraction of a specialist's hourly wage (~$50–80/hour loaded), making automation cost-favorable by 10–20× for routine documentation tasks. Setup and ongoing refinement add overhead, but the long-term ratio strongly favors AI once integrated.
Cost vs. human wageclaude-sonnet-54/5Generating and updating documentation via AI costs a small fraction of an engineer's hourly wage, though integration with network monitoring tools and review adds some overhead.
Technical feasibility todayclaude-haiku-4-5-202510013/5Tools like GPT-4, Claude, and specialized documentation platforms can generate technical documentation reliably for straightforward installations, but struggle with complex edge cases, non-standard architectures, and security-sensitive details. Products exist and see real use, but material error rates persist in production and require professional oversight.
Technical feasibility todayclaude-sonnet-53/5Products like Copilot, ChatGPT, and specialized IT documentation tools are used in production to draft network documentation, but accuracy on complex topologies and change tracking still requires human verification.

Test computer software or hardware, using standard diagnostic testing equipment and procedures.

59

CI 38–81 · exposure 50 · augmentation 88 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5IT and software development sectors have rapidly and deeply adopted automated testing and CI/CD pipelines, with major technology companies and enterprises running continuous automated diagnostics in production as standard practice.
Sector adoption velocityclaude-sonnet-53/5IT operations widely use automated monitoring and diagnostic tools, but full automation of hands-on testing remains uneven across organizations of varying digitization levels.
Augmentation potentialclaude-haiku-4-5-202510015/5AI-driven testing tools significantly augment human specialists by automatically running comprehensive test suites, flagging anomalies, and generating detailed reports, freeing support staff to focus on interpreting results and troubleshooting complex failures rather than manual test execution.
Augmentation potentialclaude-sonnet-54/5AI-driven diagnostic and monitoring tools significantly speed up issue identification and pattern recognition, letting technicians focus on remediation rather than manual detection.
Task automatabilityclaude-haiku-4-5-202510014/5Standard diagnostic testing procedures are highly structured and repetitive, involving running predefined test suites and analyzing results against known criteria. AI systems can execute most of these tests and compare outputs to benchmarks, achieving well over 50% time savings, though edge cases and novel failure modes may require human intervention.
Task automatabilityclaude-sonnet-52/5Some diagnostic test execution can be scripted or automated with monitoring tools, but hands-on hardware testing, physical connections, and interpreting ambiguous results still require human judgment and physical presence.
Adoption barriersclaude-haiku-4-5-202510012/5Few regulatory or legal barriers exist; organizations can run automated diagnostics without licensed personnel approval. Adoption friction comes mainly from integration overhead and organizational preference to retain human oversight, not hard legal requirements.
Adoption barriersclaude-sonnet-52/5No strict licensing requirement, but organizational reliance on trained IT staff for judgment calls and physical access to equipment creates moderate friction.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated diagnostic testing tools incur minimal inference and compute costs compared to paying a support specialist's loaded hourly wage for the same volume of routine tests, easily achieving order-of-magnitude cost advantages at scale.
Cost vs. human wageclaude-sonnet-52/5Diagnostic software tools are relatively cheap to run, but physical hardware testing and complex troubleshooting still require a skilled technician, keeping overall costs comparable to human labor.
Technical feasibility todayclaude-haiku-4-5-202510014/5Mature products exist for automated software testing (e.g., Selenium, Cypress, AI-powered test frameworks) and hardware diagnostics that run at scale in production environments. These tools reliably execute standard diagnostic procedures, though complex or novel hardware scenarios still show material error rates.
Technical feasibility todayclaude-sonnet-52/5Network monitoring and diagnostic software exist and are widely deployed, but they assist rather than fully replace human-run diagnostic testing procedures, especially for hardware issues.

Install network software, including security or firewall software.

59

CI 41–76 · exposure 58 · augmentation 75 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510015/5IT and cloud-native sectors have rapidly and deeply adopted automated deployment and configuration management tools; DevOps practices and Infrastructure-as-Code are now standard in enterprise and SaaS environments, with widespread production deployment.
Sector adoption velocityclaude-sonnet-53/5IT/network operations show middling AI adoption—automation and AIOps tools are increasingly piloted, but full autonomous handling of security-critical installs remains uncommon in production.
Augmentation potentialclaude-haiku-4-5-202510014/5AI and automation tools significantly augment human network specialists by handling routine installations and configurations, freeing them to focus on complex troubleshooting, security validation, and architectural decisions while remaining in the oversight loop.
Augmentation potentialclaude-sonnet-54/5AI significantly assists specialists by generating configuration scripts, explaining firewall rules, drafting documentation, and diagnosing errors, meaningfully speeding up the human-led installation process.
Task automatabilityclaude-haiku-4-5-202510014/5Installing network software can be largely automated through infrastructure-as-code, configuration management tools (Ansible, Puppet, Chef), and scripted deployments that current systems handle end-to-end, achieving substantial time savings over manual installation and configuration.
Task automatabilityclaude-sonnet-53/5AI can generate installation scripts, configuration templates, and troubleshoot common issues, but physical/remote hands-on setup, environment-specific decisions, and verification still typically require human execution and judgment., so only partial time savings are realized end-to-end.
Adoption barriersclaude-haiku-4-5-202510013/5Some organizational friction exists around change management, compliance verification, and audit trails that may require human sign-off; however, no hard licensing barrier mandates human installation itself, though security policies may impose procedural oversight requirements.
Adoption barriersclaude-sonnet-53/5No licensing requirement typically exists, but organizational risk aversion around security infrastructure, need for accountability/sign-off, and liability concerns for firewall misconfiguration create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510015/5Automated deployment and configuration tools cost orders of magnitude less than human labor when amortized across multiple installations, with minimal marginal cost per additional system once pipelines are established.
Cost vs. human wageclaude-sonnet-52/5While AI can cut some scripting/documentation time cheaply, the overall task still requires skilled human oversight, physical access, and validation, keeping all-in costs closer to comparable rather than order-of-magnitude cheaper.
Technical feasibility todayclaude-haiku-4-5-202510014/5Mature products and cloud platforms (AWS, Azure, Kubernetes, enterprise configuration management systems) reliably perform software installation and deployment at scale in production environments today, though edge cases and legacy systems may require human oversight.
Technical feasibility todayclaude-sonnet-52/5AI coding/config assistants and IT automation tools exist and are used to generate scripts and configs, but no mature product autonomously installs and validates network security/firewall software in production without a human network engineer.

Train users in procedures related to network applications software or related systems.

56

CI 41–70 · exposure 45 · augmentation 75 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Corporate IT and tech-heavy sectors are rapidly adopting AI-powered learning platforms, microlearning, and automated onboarding systems; evidence of production deployment in Fortune 500 companies is substantial.
Sector adoption velocityclaude-sonnet-53/5IT departments increasingly use AI chatbots and self-service knowledge bases for training support, a middling but growing adoption pattern in tech-forward sectors.
Augmentation potentialclaude-haiku-4-5-202510014/5AI augments trainers by generating personalized learning paths, auto-grading assessments, providing real-time knowledge bases, and handling initial Q&A, freeing trainers to focus on mentorship and complex problem-solving.
Augmentation potentialclaude-sonnet-54/5AI can generate training documentation, FAQs, and interactive tutorials, and answer routine user questions, substantially boosting the trainer's efficiency while they remain involved for complex cases.
Task automatabilityclaude-haiku-4-5-202510014/5Most of the training delivery can be automated through AI-generated documentation, video tutorials, interactive modules, and chatbot support for common questions, reducing trainer time by >50%. However, hands-on troubleshooting and live problem-solving with complex edge cases still benefit from human interaction.
Task automatabilityclaude-sonnet-52/5Training requires live diagnosis of user knowledge gaps, adapting delivery style, and hands-on troubleshooting demonstrations that current AI cannot fully replicate end-to-end, though AI can produce training materials.
Adoption barriersclaude-haiku-4-5-202510012/5Few regulatory or licensing barriers exist for automating training delivery; organizations may prefer human trainers for relationship-building, but nothing legally prohibits AI-led training modules for network software.
Adoption barriersclaude-sonnet-52/5No licensing requirement blocks AI from assisting with training, though organizations often prefer human trainers for complex onboarding and hands-on support, creating mild friction.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-driven training platforms (inference, content generation, LMS hosting) cost a fraction of loaded trainer wages per user trained, especially at scale; one system can serve hundreds of users simultaneously.
Cost vs. human wageclaude-sonnet-53/5AI-generated training materials and chatbot support reduce some labor cost, but live training and troubleshooting still require human specialist time, keeping costs roughly comparable when quality is factored in.
Technical feasibility todayclaude-haiku-4-5-202510013/5AI products like learning management systems with AI tutors, auto-generated training content, and conversational AI exist and are deployed, but often lack sophistication in handling nuanced software-specific edge cases or adapting to individual learning styles at the level human trainers provide.
Technical feasibility todayclaude-sonnet-52/5AI chatbots and documentation generators exist for IT help content, but no deployed product reliably conducts full user training sessions on specific network applications without human facilitation.

Evaluate local area network (LAN) or wide area network (WAN) performance data to ensure sufficient availability or speed, to identify network problems, or for disaster recovery purposes.

54

CI 42–66 · exposure 47 · augmentation 75 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5IT operations and network management are high-digitization sectors with rapid adoption of AI-powered monitoring platforms. Enterprise adoption of machine-learning-based network analytics is widespread and accelerating in cloud-native and digital-first organizations.
Sector adoption velocityclaude-sonnet-53/5IT operations is a moderately fast-adopting sector for AIOps and monitoring tools, though full automation of diagnosis and recovery planning remains largely human-supervised in production.
Augmentation potentialclaude-haiku-4-5-202510014/5AI excels at assisting network specialists by automating data collection, anomaly detection, and alert triage, freeing humans to focus on root-cause analysis and strategic remediation. This significantly amplifies specialist productivity while keeping human judgment in control.
Augmentation potentialclaude-sonnet-54/5AI-driven monitoring and analytics tools significantly enhance a specialist's ability to detect anomalies, correlate metrics, and prioritize troubleshooting, while humans retain decision-making authority.
Task automatabilityclaude-haiku-4-5-202510013/5AI can analyze network performance data, identify patterns, and flag anomalies or bottlenecks automatically, covering perhaps 50–70% of routine monitoring work. However, contextual judgment about root causes, business impact prioritization, and disaster recovery strategy still require human expertise, preventing full end-to-end automation at quality parity.
Task automatabilityclaude-sonnet-52/5AI can help analyze network performance data and flag anomalies, but diagnosing complex network problems and planning disaster recovery still require human judgment integrating context, topology knowledge, and organizational priorities.
Adoption barriersclaude-haiku-4-5-202510012/5Network performance evaluation has low regulatory barriers and no licensing requirement for the automation itself. Organizations are free to deploy AI monitoring systems; the main friction is organizational preference to retain human oversight for critical infrastructure decisions, not legal mandate.
Adoption barriersclaude-sonnet-52/5No licensing requirement blocks AI use, but organizational risk aversion around network outages and disaster recovery planning creates moderate friction for full automation.
Cost vs. human wageclaude-haiku-4-5-202510014/5Automated network monitoring via AI-enabled platforms is substantially cheaper than hiring dedicated staff to manually review logs and performance metrics 24/7. The per-incident analysis cost for AI is orders of magnitude lower, though integration and oversight still carry some expense.
Cost vs. human wageclaude-sonnet-52/5Monitoring software reduces some labor but requires licensing, integration, and ongoing specialist oversight, so total cost savings versus a human specialist are moderate rather than order-of-magnitude cheaper.
Technical feasibility todayclaude-haiku-4-5-202510014/5Network monitoring and performance analysis tools with AI-powered anomaly detection and alerting are deployed in production across many organizations. Tools like Cisco's AI Network Analytics, Splunk, Datadog, and others perform this reliably at scale, though false-positive rates and the need for human validation remain common.
Technical feasibility todayclaude-sonnet-53/5Network monitoring products (e.g., AIOps tools like Cisco AI Network Analytics, SolarWinds) reliably flag anomalies and performance issues today, but full evaluation and root-cause diagnosis still commonly require specialist review.

Perform routine maintenance or standard repairs to networking components or equipment.

49

CI 38–61 · exposure 38 · augmentation 88 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5IT and network operations are among the early adopters of AI/automation (incident response automation, AIOps platforms). Most medium-to-large enterprises have deployed some form of automated network monitoring and remediation, especially for routine maintenance tasks.
Sector adoption velocityclaude-sonnet-53/5IT/network operations sectors are moderately fast adopters of AIOps and monitoring tools, though physical maintenance tasks lag behind software-only workflows in automation.
Augmentation potentialclaude-haiku-4-5-202510015/5AI already augments network specialists through real-time anomaly detection, automated root-cause analysis, patch deployment recommendations, and auto-remediation suggestions. These tools significantly accelerate diagnosis and reduce manual toil while the specialist retains oversight and judgment.
Augmentation potentialclaude-sonnet-54/5AI-based network monitoring, predictive maintenance alerts, and diagnostic assistants significantly help technicians identify issues faster and guide repair procedures, improving productivity while the human still performs physical work.
Task automatabilityclaude-haiku-4-5-202510013/5AI-guided systems can partially automate routine diagnostics, software configuration updates, and documentation of standard fixes, but physical repairs (cable replacement, hardware swaps) require robotic systems not widely deployed in current production. The mix of remote and hands-on work limits end-to-end automation to roughly half the task scope.
Task automatabilityclaude-sonnet-52/5This task involves physical handling of hardware (cabling, swapping components, rebooting devices, physical inspection) which current AI cannot perform; software-side diagnostics can be partially assisted but the core hands-on maintenance is not automatable end-to-end.
Adoption barriersclaude-haiku-4-5-202510012/5Few legal or licensing barriers prevent automation of diagnostics and routine updates in most jurisdictions. Some organizations require human sign-off for production changes, but this is organizational policy rather than regulatory mandate. Physical access constraints are practical rather than regulatory.
Adoption barriersclaude-sonnet-52/5No licensing requirement typically gates this work, but physical access, safety, and organizational IT policies create moderate friction against pure automation of hands-on repairs.
Cost vs. human wageclaude-haiku-4-5-202510014/5AI-driven diagnostic and maintenance tools have fallen sharply in cost relative to specialist labor. The remote/monitoring portions can be handled at a fraction of a technician's fully loaded wage, though the physical intervention component still requires human labor.
Cost vs. human wageclaude-sonnet-52/5While software diagnostics can be cheap to run, physical repair still requires technician labor, travel, and tools, so overall cost is not substantially reduced by AI relative to the human-performed portions.
Technical feasibility todayclaude-haiku-4-5-202510013/5Deployed products exist for automated network diagnostics, patch management, and ticket triage (e.g., Splunk, Datadog integrations with AI), but these handle the knowledge and remote work layers; hardware repair automation remains research-stage. Error tolerance is high for configuration changes but lower for physical repairs.
Technical feasibility todayclaude-sonnet-52/5AI-driven network monitoring and diagnostic tools are deployed in production for anomaly detection and alerting, but actual physical maintenance/repair still requires a human technician on-site, so no product performs the full task reliably.

Troubleshoot network or connectivity problems for users or user groups.

44

CI 32–55 · exposure 38 · augmentation 75 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Tech companies and large enterprises use AI-assisted ticketing and log analysis tools, but deployment is primarily for triage and augmentation rather than replacement. Smaller organizations and legacy sectors lag, and pure automation adoption remains low due to reliability concerns and business continuity risk.
Sector adoption velocityclaude-sonnet-53/5IT support functions are adopting AI chatbots and automated diagnostics steadily, but most organizations still rely on human specialists for anything beyond basic tier-1 troubleshooting.
Augmentation potentialclaude-haiku-4-5-202510014/5AI provides strong productivity gains through log parsing, symptom matching against knowledge bases, and automated diagnostic scripts that help specialists narrow scope and prioritize steps. AI copilots for ticket resolution and real-time guidance materially reduce time to fix while maintaining human oversight.
Augmentation potentialclaude-sonnet-54/5AI significantly assists specialists by summarizing logs, suggesting root causes, and automating repetitive diagnostic steps, letting humans focus on complex resolution and physical remediation.
Task automatabilityclaude-haiku-4-5-202510012/5Troubleshooting network problems involves diagnosing highly variable, context-dependent issues that often require physical inspection, user interaction, and real-time system analysis. While AI can assist with log analysis and common-cause diagnosis, end-to-end troubleshooting to resolution with 50% time savings requires human judgment, access to live systems, and adaptive problem-solving that current AI cannot reliably handle independently.
Task automatabilityclaude-sonnet-53/5AI can automate diagnosis of common connectivity issues (DNS, DHCP, config errors) via log analysis and scripted remediation, but complex or novel network problems requiring physical inspection or ambiguous symptom interpretation still need human judgment.
Adoption barriersclaude-haiku-4-5-202510013/5While there are no strict licensing barriers, IT support roles carry organizational accountability for system availability and data security. Regulatory compliance (HIPAA, SOC 2) and customer trust requirements create friction against full automation, though augmentation is permitted and common.
Adoption barriersclaude-sonnet-52/5No licensing requirement exists for this role, but organizational trust, need for physical access to hardware, and liability for network downtime create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Network troubleshooting requires real-time system access, specialized domain knowledge, and often physical intervention. AI-assisted tools reduce overhead but do not yet achieve cost parity with human specialists; oversight and validation costs remain material, and full automation is not viable.
Cost vs. human wageclaude-sonnet-53/5AI-driven triage and diagnostics reduce ticket-handling time and cost, but licensing for monitoring platforms plus required human escalation for complex issues keeps costs roughly comparable to skilled technician time for full resolution.
Technical feasibility todayclaude-haiku-4-5-202510012/5Products exist for log analysis and ticket triage (e.g., chatbots), but no deployed system reliably troubleshoots and resolves connectivity issues end-to-end in production environments. Most implementations require human specialists to validate findings and execute fixes, particularly for complex or novel failures.
Technical feasibility todayclaude-sonnet-53/5AIOps and network monitoring tools (e.g., Cisco AI Network Analytics, chatbots for tier-1 helpdesk triage) are deployed in production, but they mostly handle known-pattern issues and still escalate complex cases to humans.

Provide telephone support related to networking or connectivity issues.

42

CI 25–59 · exposure 38 · augmentation 63 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption in this domain remains slow; most organizations use basic IVR for routing and knowledge bases as adjuncts, but genuine AI-driven support at scale remains uncommon, with heavy reliance on live specialist phone support as the standard.
Sector adoption velocityclaude-sonnet-53/5IT support functions are moderately digitized with growing chatbot/AI triage adoption, but many organizations still rely heavily on human phone support for network issues, placing this at a middling adoption pace.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can assist specialists by suggesting diagnostics, retrieving relevant documentation, and automating simple preliminary steps, improving human productivity on portions of the task, though the core troubleshooting interaction still requires human judgment and expertise.
Augmentation potentialclaude-sonnet-54/5AI tools significantly assist support specialists via automated diagnostics, knowledge-base retrieval, and call transcription/summarization, boosting productivity while humans remain in the loop for complex troubleshooting.
Task automatabilityclaude-haiku-4-5-202510012/5AI can partially help with basic troubleshooting scripts and issue categorization, but complex network diagnostics require real-time system access, nuanced understanding of customer environments, and interactive problem-solving that falls short of the 50% time-saving threshold for full task automation.
Task automatabilityclaude-sonnet-53/5Chatbots and voice AI can resolve common connectivity issues (password resets, basic troubleshooting scripts) but complex network diagnostics still require human expertise and escalation, so only a portion of calls meet the 50% time-saving bar.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory and organizational barriers are moderate-to-high: many enterprises require human accountability for network support, liability concerns over misdiagnosis, and contractual requirements that support interactions be documented and verifiable by qualified human specialists.
Adoption barriersclaude-sonnet-52/5No licensing requirement exists for this work, though some organizations maintain human support lines for customer trust and complex issue resolution, creating moderate but not hard barriers.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI infrastructure (speech recognition, LLM inference, oversight) combined with integration and human review overhead makes the all-in cost comparable to or exceeding a support specialist's loaded wage, especially for complex cases requiring escalation.
Cost vs. human wageclaude-sonnet-54/5Automated phone/chat support systems are much cheaper per interaction than a live technician for routine issues, though complex cases still require costlier human intervention, keeping it just below the top tier.
Technical feasibility todayclaude-haiku-4-5-202510012/5While chatbots and IVR systems exist for initial triage, no deployed product reliably handles end-to-end telephone support for networking issues without significant human handoff; audio understanding and real-time diagnostic interaction remain error-prone at scale.
Technical feasibility todayclaude-sonnet-53/5AI-driven IT helpdesk tools (e.g., automated triage bots, virtual assistants) are deployed in production for tier-1 support, but they handle a narrow scope and frequently escalate to human specialists for anything beyond scripted fixes.

Install new hardware or software systems or components, ensuring integration with existing network systems.

39

CI 30–49 · exposure 38 · augmentation 75 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510014/5Large enterprises and cloud-native organizations have adopted Infrastructure-as-Code and automated deployment pipelines extensively; smaller firms and legacy-heavy sectors lag, but overall velocity across the tech sector is high with growing DevOps/SRE practices.
Sector adoption velocityclaude-sonnet-52/5IT infrastructure/network operations is a moderate-adoption sector; automation tools like Ansible/Terraform exist but AI-driven autonomous deployment is still nascent and mostly pilot-stage.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-assisted configuration management, automated pre-flight checks, and intelligent deployment planning significantly enhance specialist productivity by automating tedious steps, freeing humans to focus on integration validation and troubleshooting, creating a strong augmentation dynamic.
Augmentation potentialclaude-sonnet-54/5AI can generate configuration scripts, troubleshoot compatibility issues, draft documentation, and suggest integration steps, meaningfully speeding up the human-led installation process.
Task automatabilityclaude-haiku-4-5-202510013/5Installation of routine hardware/software components can be substantially automated (driver deployment, configuration scripts, testing); however, ensuring integration with diverse legacy systems and handling edge cases requires human oversight, limiting time savings to roughly 50% on well-standardized deployments.
Task automatabilityclaude-sonnet-52/5Physical hardware installation and much system integration requires hands-on work, troubleshooting unique environments, and coordination that current AI cannot fully execute end-to-end.assign only scripting/config portions are automatable.
Adoption barriersclaude-haiku-4-5-202510013/5Organizational friction and risk aversion are moderate: IT teams prefer human sign-off on critical network changes for liability and accountability reasons, and many organizations mandate human testing and validation before deployment to production systems.
Adoption barriersclaude-sonnet-53/5No licensing requirement typically, but organizational risk aversion around network changes, security policies, and change-management approval processes create meaningful friction.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI/automation tooling reduces manual overhead significantly, but comprehensive integration testing, troubleshooting unexpected incompatibilities, and human oversight still require skilled network staff, keeping total costs comparable to or only moderately below direct human labor.
Cost vs. human wageclaude-sonnet-52/5Physical installation, cabling, and troubleshooting still require paid technician time on-site; AI tools reduce some planning/config time but don't eliminate the bulk of labor cost.
Technical feasibility todayclaude-haiku-4-5-202510013/5Products like infrastructure-as-code platforms (Ansible, Terraform) and CI/CD systems reliably automate parts of software installation in controlled environments, but hardware installation and integration testing with heterogeneous networks remain error-prone and typically require human validation in production.
Technical feasibility todayclaude-sonnet-52/5AI-assisted configuration tools and scripts exist, but no deployed product autonomously installs and integrates new hardware/software into a live network without significant human execution and oversight.

Configure and define parameters for installation or testing of local area network (LAN), wide area network (WAN), hubs, routers, switches, controllers, multiplexers, or related networking equipment.

37

CI 25–50 · exposure 33 · augmentation 63 · importance 3.9/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5While IT organizations use AI-assisted tools for some documentation and minor configuration tasks, production network setup remains largely manual and specialist-driven due to the high cost of misconfiguration. Adoption of full automation is slow in conservative enterprise environments.
Sector adoption velocityclaude-sonnet-53/5IT/network operations sectors show moderate AI tool adoption (config generators, network automation platforms) but full agentic automation of network configuration in production is still limited.
Augmentation potentialclaude-haiku-4-5-202510013/5AI can meaningfully assist by suggesting parameter values, auto-generating configuration templates, and identifying common errors in configs, helping specialists work faster. However, the human specialist remains necessary for validation, testing, and judgment, making this a useful but bounded augmentation.
Augmentation potentialclaude-sonnet-54/5AI tools substantially help network specialists draft configurations, troubleshoot syntax, and explain parameter options, meaningfully speeding up the human-led configuration process.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with configuration syntax and parameter selection, this task requires hands-on equipment setup, real-time testing feedback, and environment-specific customization that current AI cannot reliably execute end-to-end without human intervention. The physical and contextual complexity prevents the 50% time-saving threshold.
Task automatabilityclaude-sonnet-53/5AI can generate configuration scripts and templates (e.g., Cisco IOS commands) and assist with parameter selection, but physical installation, environment-specific validation, and troubleshooting still require significant human judgment and hands-on work.
Adoption barriersclaude-haiku-4-5-202510014/5Network configuration often requires licensed credentials (CompTIA Network+, vendor certifications), organizational accountability for infrastructure stability, and liability for downtime. Many enterprises mandate human sign-off on production network changes for compliance and risk reasons.
Adoption barriersclaude-sonnet-52/5No formal licensing requirement is typical, but organizational risk aversion around network downtime and security means human sign-off is standard practice before changes go live.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted configuration tools reduce some manual work, but the specialist's expertise in troubleshooting, validation, and environment-specific tuning remains essential and costly. AI integration costs plus human oversight typically do not undercut specialist labor significantly.
Cost vs. human wageclaude-sonnet-53/5AI can reduce time spent drafting configurations, but oversight, testing, and physical installation costs remain, so overall savings are moderate rather than order-of-magnitude cheaper.
Technical feasibility todayclaude-haiku-4-5-202510012/5AI tools can suggest configurations and parameters based on documentation, but no deployed product reliably configures live network equipment autonomously. Manual verification, device interaction, and debugging of actual hardware remain necessary, limiting current production viability.
Technical feasibility todayclaude-sonnet-52/5Network automation tools and AI-assisted config generators exist, but production deployments still rely heavily on human network engineers to verify, test, and adapt configs to specific hardware and topologies.

Configure security settings or access permissions for groups or individuals.

36

CI 28–45 · exposure 38 · augmentation 75 · importance 4.4/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Large enterprises in high-compliance sectors (finance, healthcare, tech) are adopting AI-assisted IAM and policy automation, but small-to-medium organizations lag, and many still rely on manual review; adoption is uneven and not yet dominant in the sector.
Sector adoption velocityclaude-sonnet-53/5IT/network administration is a moderately digitized field with growing use of AI-driven IT service management tools, but adoption for security-critical configuration remains cautious and pilot-stage in many organizations.
Augmentation potentialclaude-haiku-4-5-202510014/5AI substantially augments this task by auto-generating policy templates, flagging permission drift, suggesting least-privilege configurations, and validating rule sets, meaningfully accelerating the specialist's workflow while they retain responsibility for approval and deployment.
Augmentation potentialclaude-sonnet-54/5AI tools (e.g., IT automation platforms, natural-language policy generators) significantly speed up drafting and auditing of access rules while a human technician still approves and applies changes.
Task automatabilityclaude-haiku-4-5-202510013/5AI can assist with generating access control policies, flagging misconfigurations, and automating routine permission provisioning for standard use cases, but complex organizational context, exception handling, and security risk assessment still require human judgment and are not routinely automated end-to-end today.
Task automatabilityclaude-sonnet-52/5This requires understanding organizational context, risk assessment, and judgment about appropriate access levels, which AI can assist with but not fully own end-to-end without human validation.
Adoption barriersclaude-haiku-4-5-202510014/5Security and compliance regulations (SOC 2, HIPAA, GDPR) typically mandate human accountability, documented approval chains, and auditable sign-offs for access changes; liability and legal risk create strong barriers to unattended automation of permission settings.
Adoption barriersclaude-sonnet-54/5Security configuration often requires accountability, audit trails, and sometimes compliance sign-off (e.g., SOC2, HIPAA), creating liability and governance barriers to full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5While AI-driven identity and access management (IAM) platforms can reduce overhead, they still require infrastructure investment, training, and human specialists to validate and oversee configurations; the all-in cost remains comparable to or higher than a skilled operator for most organizations.
Cost vs. human wageclaude-sonnet-53/5AI-assisted scripting and automation tools reduce time spent on repetitive permission tasks, but human oversight and validation costs keep overall cost roughly comparable to a skilled technician's time for critical configurations.
Technical feasibility todayclaude-haiku-4-5-202510013/5Tools exist (identity management platforms, policy-as-code frameworks, security scanning) that partially automate access rule generation and validation, but they require significant manual configuration, review, and sign-off; no mature product fully automates the task from requirements to deployment without human oversight.
Technical feasibility todayclaude-sonnet-52/5Some IT automation tools and AI copilots can suggest or apply permission templates, but reliable autonomous configuration of security settings in production networks is rare due to error/risk sensitivity.

Test repaired items to ensure proper operation.

34

CI 30–37 · exposure 30 · augmentation 63 · importance 3.6/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5IT departments have adopted automated testing frameworks and continuous integration/monitoring tools at a steady pace, but deep, autonomous testing deployment remains incomplete; many organizations still rely on hybrid manual-plus-tool testing for high-stakes network repairs.
Sector adoption velocityclaude-sonnet-52/5IT support functions are adopting AI diagnostic and monitoring tools at a moderate pace, but full testing automation for repaired items remains uncommon in production settings.
Augmentation potentialclaude-haiku-4-5-202510014/5AI-driven diagnostic and monitoring tools significantly augment technician productivity by automatically running test suites, identifying anomalies, and flagging issues for human review, allowing technicians to focus on interpreting results and making corrections rather than executing rote checks.
Augmentation potentialclaude-sonnet-53/5AI-based network diagnostic tools and scripts can assist technicians in verifying repairs by running automated checks and flagging anomalies, improving efficiency while human oversight remains essential.
Task automatabilityclaude-haiku-4-5-202510012/5Testing repaired network items requires interpreting complex, context-dependent outcomes and making judgments about whether behavior is acceptable; while basic connectivity checks can be scripted, the full task of validating proper operation across configurations demands human expertise and decision-making that current AI cannot reliably handle end-to-end.
Task automatabilityclaude-sonnet-52/5Testing repaired network hardware/software often requires physical access, plugging in devices, observing lights, running diagnostics in context-specific environments, which current AI cannot fully perform end-to-end.'
Adoption barriersclaude-haiku-4-5-202510013/5Testing is often part of a broader troubleshooting workflow and organizational change-management processes; however, there are no strict licensing or legal mandates requiring a human to perform the testing itself, only informal preferences for human sign-off on critical infrastructure.
Adoption barriersclaude-sonnet-53/5No strict licensing requirement, but organizational risk tolerance and need for physical verification of hardware/network functionality create moderate friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Diagnostic and testing tools are relatively affordable, but the setup, integration with legacy systems, and human oversight required to validate results reliably still approaches or exceeds the loaded cost of a technician performing focused testing manually.
Cost vs. human wageclaude-sonnet-52/5Physical verification and hands-on testing still requires human presence and judgment, so AI tools reduce but don't eliminate the labor cost, keeping costs roughly comparable to human-only workflows.
Technical feasibility todayclaude-haiku-4-5-202510013/5Automated testing tools and diagnostic software exist and are in production use (e.g., network monitoring platforms, packet analyzers), but they typically require human interpretation of results and judgment calls about edge cases, false positives, and acceptable performance thresholds.
Technical feasibility todayclaude-sonnet-52/5Some automated network monitoring and diagnostic tools exist, but they assist rather than autonomously perform post-repair verification testing in production environments.

Analyze and report computer network security breaches or attempted breaches.

32

CI 28–37 · exposure 30 · augmentation 75 · importance 4.3/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Large financial services, tech, and healthcare organizations have adopted AI/ML-driven threat detection in production; mid-market firms pilot SIEM+AI. However, smaller organizations lag, and the sector overall still relies heavily on human analysts for validation and judgment. Adoption is accelerating but not yet deep across the sector.
Sector adoption velocityclaude-sonnet-53/5Cybersecurity is a moderately fast-adopting field with many AI-augmented SOC tools in pilot and production use, though full autonomous incident response remains uncommon.
Augmentation potentialclaude-haiku-4-5-202510014/5AI dramatically augments security analysts by automating log aggregation, alert triage, correlation of events across systems, and rapid pattern matching against threat intelligence databases. Analysts can focus on investigation, remediation, and attribution rather than manual log review, substantially raising their throughput and effectiveness.
Augmentation potentialclaude-sonnet-54/5AI significantly enhances analyst productivity by automating log correlation, flagging anomalies, and drafting summary reports, while humans retain oversight for validation and response decisions.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can assist with detecting anomalies in network logs and flagging suspicious patterns, analyzing breach severity, attribution, and business impact requires contextual judgment and investigation of novel attack vectors. Current systems excel at pattern matching within known threat categories but struggle with sophisticated, novel, or multi-stage breaches that demand human expertise.
Task automatabilityclaude-sonnet-52/5AI can assist with log analysis, anomaly detection, and drafting incident reports, but investigating breaches requires contextual judgment, correlating diverse evidence, and making determinations that current systems cannot fully replace end-to-end.
Adoption barriersclaude-haiku-4-5-202510014/5Regulatory frameworks (HIPAA, GDPR, PCI-DSS, SOC 2) typically mandate that breach investigations and formal incident reporting be conducted by qualified personnel, often with audit trails and accountability. Liability and legal exposure for missed or misreported breaches creates strong organizational and compliance barriers to full automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement mandates a human, but liability, compliance reporting obligations (e.g., breach disclosure laws), and the high cost of an incorrect security determination create meaningful organizational friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5Security specialists command high loaded wages (salary, benefits, training). While SIEM/AI monitoring reduces per-task cost compared to pure manual review, the specialist must still interpret alerts, investigate context, and certify findings. Integration and tuning of detection systems adds overhead, keeping total cost-per-breach-analysis close to human-competitive.
Cost vs. human wageclaude-sonnet-52/5AI tools reduce alert volume and triage time but still require skilled human analysts for investigation and reporting, so overall cost savings are moderate rather than transformative.
Technical feasibility todayclaude-haiku-4-5-202510013/5Security information and event management (SIEM) products and AI-driven threat detection systems are widely deployed and can identify many classes of breaches automatically. However, they produce false positives, miss sophisticated attacks, and require human analysts to validate findings, reconstruct attack chains, and assess real impact—hence production systems are reliable only with human oversight.
Technical feasibility todayclaude-sonnet-52/5SIEM/SOAR platforms with AI-driven anomaly detection and alert triage are deployed in production, but human analysts still perform the core investigation and reporting due to high false-positive rates and complex attack patterns.

Configure wide area network (WAN) or local area network (LAN) routers or related equipment.

31

CI 25–37 · exposure 30 · augmentation 75 · importance 4.0/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510013/5Mid-market to enterprise IT environments are adopting AI-assisted network management tools in pilots and some production deployments, but adoption remains cautious and incomplete. Legacy systems, risk aversion, and the need for human expertise slow widespread deep adoption.
Sector adoption velocityclaude-sonnet-52/5IT infrastructure and networking sectors are moderately digitized but adoption of AI-driven network configuration remains in early pilot stages compared to faster-adopting domains like software coding or customer support.
Augmentation potentialclaude-haiku-4-5-202510014/5AI significantly assists specialists by auto-generating candidate configurations, validating syntax, suggesting optimizations, and flagging inconsistencies. These tools measurably boost productivity when specialists retain control and judgment over deployment decisions.
Augmentation potentialclaude-sonnet-54/5AI tools can meaningfully speed up drafting configuration scripts, documentation, and troubleshooting suggestions, letting network specialists focus on validation, physical setup, and complex judgment calls.
Task automatabilityclaude-haiku-4-5-202510012/5While AI can help generate router configurations and troubleshoot common issues, end-to-end autonomous configuration requires understanding complex network topology, security policies, and business requirements that typically demand human judgment and validation. Current AI cannot reliably handle the full decision tree without expert oversight.
Task automatabilityclaude-sonnet-52/5WAN/LAN router configuration involves physical access, site-specific topology decisions, and troubleshooting hardware quirks that current AI cannot fully handle end-to-end, though AI can generate config scripts and suggest settings.
Adoption barriersclaude-haiku-4-5-202510014/5Strong barriers exist: network changes carry high risk of downtime and security exposure, requiring licensed/certified specialists to review and sign off in most organizations. Liability, regulatory compliance, and customer trust in human expertise create substantial friction against full automation.
Adoption barriersclaude-sonnet-53/5No licensing requirement, but organizational risk aversion around network outages, security implications of misconfiguration, and need for physical presence create meaningful friction against full automation.
Cost vs. human wageclaude-haiku-4-5-202510012/5AI-assisted configuration tools reduce labor but require initial licensing, integration, and a support specialist to validate, architect, and troubleshoot failures. The total cost including human oversight often approaches or exceeds the cost of direct specialist labor, especially for complex non-standard networks.
Cost vs. human wageclaude-sonnet-52/5Human network specialists remain necessary for physical installation, testing, and edge-case troubleshooting, so AI only reduces part of the labor cost while oversight and verification costs remain significant.
Technical feasibility todayclaude-haiku-4-5-202510013/5Products exist (e.g., Cisco DNA Center, Juniper Apstra) that can automate some router configuration and validation, but they require significant setup, human architecture input, and ongoing oversight. They work well for standardized environments but struggle with novel or legacy hybrid setups.
Technical feasibility todayclaude-sonnet-52/5AI copilots and automation scripts (e.g., Ansible with LLM assistance) exist for generating configs, but reliable autonomous deployment and validation on live network equipment in production is still narrow and requires human verification.

Install and configure wireless networking equipment.

30

CI 25–35 · exposure 25 · augmentation 50 · importance 3.5/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510012/5Adoption of AI for wireless installation remains low. Most organizations rely on traditional technicians or Managed Service Providers for physical deployment. While configuration automation tools are emerging in IT operations, actual hands-on installation work is not yet displaced at production scale.
Sector adoption velocityclaude-sonnet-52/5IT infrastructure and network deployment sectors have adopted AI for monitoring and troubleshooting but physical installation and configuration tasks remain largely manual, showing slow adoption of full automation.
Augmentation potentialclaude-haiku-4-5-202510013/5AI provides moderate augmentation through configuration templates, automated troubleshooting guides, documentation lookup, and signal modeling tools. However, augmentation is limited because on-site diagnosis and physical problem-solving still depend on technician judgment and hands-on adjustment.
Augmentation potentialclaude-sonnet-53/5AI-driven network configuration tools, templates, and troubleshooting assistants can meaningfully speed up the configuration portion of the task, though physical installation remains unaided.
Task automatabilityclaude-haiku-4-5-202510012/5Installing and configuring wireless equipment requires physical placement, cable routing, and hands-on device manipulation that current AI cannot perform. While AI can assist with configuration steps (scripts, documentation), the core installation task—physically mounting hardware, testing signal propagation, managing site-specific constraints—remains manual and requires on-site presence.
Task automatabilityclaude-sonnet-52/5Physical installation of wireless access points, running cables, and mounting hardware requires manual, on-site work that current AI cannot perform; configuration steps can be partially scripted but overall task retains significant hands-on component.
Adoption barriersclaude-haiku-4-5-202510014/5Significant regulatory and organizational barriers exist: many installations must comply with FCC/RF safety standards and may require licensed technician certification in some jurisdictions. Liability for network uptime, security, and safety compliance creates strong pressure for human sign-off and accountability on critical infrastructure installations.
Adoption barriersclaude-sonnet-52/5No licensing requirement typically gates this work, but organizational reliance on physical presence and network security sign-off creates some friction, though not a hard legal barrier.
Cost vs. human wageclaude-haiku-4-5-202510012/5Current AI deployment for this task (remote guidance, documentation automation, troubleshooting support) still requires technician labor for the physical work. Integration and oversight costs are substantial relative to labor savings, making the all-in cost comparable to or exceeding direct human labor.
Cost vs. human wageclaude-sonnet-52/5Physical installation still requires a technician on-site, and AI configuration tools add cost/integration overhead without eliminating labor, so all-in cost is comparable to or higher than using a technician alone for this bundled task.
Technical feasibility todayclaude-haiku-4-5-202510012/5No deployed AI system performs end-to-end wireless installation autonomously. AI tools exist for configuration guidance and troubleshooting, but real-world installation involves environmental assessment, physical problem-solving, and site-specific adaptation that current products handle only in narrow, scripted scenarios.
Technical feasibility todayclaude-sonnet-52/5No deployed product autonomously installs and configures wireless networking equipment end-to-end; existing tools (network automation platforms, AI-assisted configuration generators) only handle a slice of the configuration step under human supervision.

Install or repair network cables, including fiber optic cables.

7

CI 5–10 · exposure 0 · augmentation 25 · importance 3.7/5 · click for rater detail

Panel ratings per dimension for this task
DimensionModelRatingRationale
Sector adoption velocityclaude-haiku-4-5-202510011/5The physical and site-specific nature of cable work means even highly digitized IT departments cannot meaningfully automate this task; human technicians remain essential.
Sector adoption velocityclaude-sonnet-51/5Cable installation and repair work is physical, low-digitization field labor with essentially no AI/robotic adoption in production settings.
Augmentation potentialclaude-haiku-4-5-202510012/5AI can assist with diagnostic information, routing planning, or documentation, but the core physical task of installation and repair requires human skill and presence; the assistance is marginal relative to the task's core demands.
Augmentation potentialclaude-sonnet-52/5AI can help with documentation, troubleshooting guidance, or diagnostic support before/after the physical work, but offers minimal assistance to the hands-on cabling task itself.
Task automatabilityclaude-haiku-4-5-202510011/5This task involves physical manipulation of cables and hardware in real-world environments, requiring dexterity, spatial reasoning, and troubleshooting that current AI systems cannot perform. No end-to-end automation of cable installation or repair is feasible with today's general-purpose systems.
Task automatabilityclaude-sonnet-51/5This is a physical, hands-on task requiring running cable through buildings, terminating connectors, and fusion splicing fiber, none of which current AI systems can perform without robotic embodiment.
Adoption barriersclaude-haiku-4-5-202510014/5Physical presence on-site is a hard requirement; regulatory and liability considerations often mandate human technician sign-off for network infrastructure. Safety standards and certification further protect this work.
Adoption barriersclaude-sonnet-53/5No licensing typically required, but physical access to buildings, safety practices around fiber optics, and organizational reliance on trained technicians create moderate practical barriers to any automated substitute.
Cost vs. human wageclaude-haiku-4-5-202510011/5Specialized robots capable of cable installation would be significantly more expensive than the loaded wage of a technician, and would require extensive setup and environmental adaptation.
Cost vs. human wageclaude-sonnet-51/5There is no AI substitute for the physical labor involved, so the human technician remains the only option and cost comparison is not applicable/AI is not cheaper.
Technical feasibility todayclaude-haiku-4-5-202510011/5No deployed AI product reliably performs physical cable installation or repair work. Robotics for this domain remain specialized, expensive, and not broadly available in production environments.
Technical feasibility todayclaude-sonnet-51/5No deployed AI product installs or repairs physical network cabling; this remains entirely a manual skilled-trade activity performed by technicians.

Related occupations — Computer & Mathematical

How to read this

A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.

What would change this score

New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.