Switchboard Operators, Including Answering Service
43-2011.00Median wage $38,630/yr34,280 employed (US)Rank #16 of 923 scored · top 2% by substitution
Operate telephone business systems equipment or switchboards to relay incoming, outgoing, and interoffice calls. May supply information to callers and record messages.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution69
Exposure67
Augmentation54
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
19 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
68%
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%69
panel mean rating 3.8/5 → substitution pressure 69/100
Technical feasibility todayw 20%65
panel mean rating 3.6/5 → substitution pressure 65/100
Cost vs. human wagew 15%78
panel mean rating 4.1/5 → substitution pressure 78/100
Adoption barriersw 20%inverted — strong barriers lower the score72
panel mean rating 2.1/5 (barrier strength) → substitution pressure 72/100
Sector adoption velocityw 10%62
panel mean rating 3.5/5 → substitution pressure 62/100
Task breakdown (19 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.
Stamp messages with time and date and file them appropriately.
99CI 97–100 · exposure 100 · augmentation 25 · importance 4.1/5 · click for rater detail
Stamp messages with time and date and file them appropriately.
99| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Switchboard operations and call centers operate in highly digitized sectors (information, customer service) with rapid adoption of automation and RPA solutions already evident in industry practice. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Office/telecom administrative functions have broadly adopted automated logging and digital record systems for years, though switchboard-specific roles are a declining niche. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | While AI could assist by suggesting file destinations or auto-correcting metadata, the task itself is so simple and standardized that augmentation adds minimal value compared to full automation. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Since the task is already largely automated end-to-end, there's little marginal augmentation value for a human performing it manually alongside AI. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Stamping messages with time/date and filing them is a purely mechanical, rule-based task that modern document management and workflow automation systems routinely perform end-to-end with high speed and accuracy, easily exceeding 50% time savings at equal or better quality. |
| Task automatability | claude-sonnet-5 | 5/5 | This is a simple, structured, repetitive data-entry/timestamping and filing task that off-the-shelf systems (automated call logging, timestamping software, digital filing) already handle fully and reliably. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | There are no licensing, legal, or liability barriers to automating this purely administrative task, and it does not require human contact or judgment. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human perform simple message timestamping and filing; it's purely administrative. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | The cost of cloud-based automation and document management systems per message processed is orders of magnitude cheaper than paying human operators for the same work, especially at volume. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated timestamping and digital filing costs a tiny fraction of a cent per message versus the loaded wage of a human operator performing manual filing. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products including RPA platforms, document management systems (DMS), and email automation tools reliably perform timestamping and filing at scale in production environments across many organizations today. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Automated call/message logging systems with timestamps and digital filing are mature, widely deployed products used at scale across telecom and office systems today. |
Keep records of calls placed and charges incurred.
97CI 95–100 · exposure 100 · augmentation 50 · importance 4.0/5 · click for rater detail
Keep records of calls placed and charges incurred.
97| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Telecommunications, customer service, and business operations sectors have already deeply adopted automated call logging and billing systems; this automation is standard across the industry rather than an emerging pilot. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Telecom and business communication sectors have long since adopted automated call logging and billing systems as standard infrastructure. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | While the task is largely fully automatable, AI-assisted review dashboards and anomaly detection can help human operators verify records and flag billing discrepancies, adding value in quality assurance and oversight roles. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Since the task is already almost fully automated, there is little role for AI to 'augment' a human performing it manually. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Call logging and charge recording are highly structured, digital tasks with clear inputs (caller, recipient, duration, destination) and outputs (call records, billing data). Current AI and integrated phone systems can automatically capture, classify, and record this information end-to-end with >50% time savings and negligible error rates compared to manual entry. |
| Task automatability | claude-sonnet-5 | 5/5 | Logging calls and charges is a structured data-capture task that is fully automated by existing telephony/PBX systems, CRM software, and call-accounting platforms without human involvement. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Regulatory compliance requirements (call recording consent, billing accuracy audits) and some customer expectations for human oversight create modest friction, but no legal requirement mandates human switchboard operators perform this logging task. Technical and organizational barriers are minimal. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing, liability, or regulatory requirement mandating a human keep these records; automated systems are the norm. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated call recording and billing integration cost pennies per call in infrastructure and negligible per-task overhead, compared to the loaded wage of a human operator performing manual entry or verification. AI is orders of magnitude cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated call logging software costs pennies per call compared to paying a human operator to manually record and tally charges. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Modern telephony systems and billing platforms have embedded automation for call logging, CDR (call detail record) capture, and charge calculation. These are mature, production-grade systems deployed across telecommunications, contact centers, and business phone services at scale. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Automatic call detail recording (CDR) and billing systems have been deployed at scale in telecom and answering services for decades, making this a mature, reliable production capability. |
Page individuals to inform them of telephone calls, using paging or interoffice communication equipment.
91CI 81–100 · exposure 87 · augmentation 38 · importance 4.4/5 · click for rater detail
Page individuals to inform them of telephone calls, using paging or interoffice communication equipment.
91| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Automated call routing and paging have been adopted deeply and rapidly across information, finance, healthcare, and professional services for decades. Switchboard operator roles have already largely been displaced by technology in modern organizations. |
| Sector adoption velocity | claude-sonnet-5 | 5/5 | Telephony and office communication systems were among the earliest and most thoroughly automated business functions, with near-universal adoption of automated paging/routing. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist human operators by pre-filtering calls, suggesting the right recipient, or auto-logging messages, improving their efficiency. However, the task itself is straightforward enough that augmentation provides moderate rather than transformative gains. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Since the task is already almost fully automated, there is little residual human task left for AI to augment in a human-in-the-loop sense. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Modern AI systems can easily automate the core task of detecting incoming calls, identifying the recipient, and triggering paging or interoffice communication via APIs or integrations with phone systems. This involves minimal variability and would easily meet the 50% time-saving threshold, though some edge cases (screening, discretionary hold decisions) may require human oversight. |
| Task automatability | claude-sonnet-5 | 5/5 | Paging someone about a call is a simple, rule-based notification task that automated PBX/paging systems and answering services have handled for decades, easily meeting the 50% time-saving bar with off-the-shelf systems. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or legal barriers exist to automating paging; no licensing is required and liability is minimal for missed pages in most organizations. The main friction is organizational inertia and customer/employee preference for human receptionists, not hard legal restrictions. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human perform simple call notification; organizations have freely automated this for years. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated call routing and paging systems cost a fraction of a full-time switchboard operator's loaded wage (often $35k–$50k annually). Once infrastructure is in place, per-call costs are negligible, making AI/automation orders of magnitude cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated paging/telephony systems cost a fraction of a cent per call versus a human operator's wage for the same notification task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed voicemail systems, call-routing software, and modern PBX systems already perform this function reliably in production at scale. AI-driven call routing and automated paging systems are standard in many organizations, though some legacy systems or complex scenarios still rely on human operators. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Automated paging, voicemail routing, and call-forwarding systems are mature, deployed at scale across offices, hospitals, and businesses today with high reliability. |
Relay or route written or verbal messages.
89CI 79–100 · exposure 87 · augmentation 50 · importance 4.3/5 · click for rater detail
Relay or route written or verbal messages.
89| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Switchboard and answering services have been undergoing deep automation for over a decade. Enterprise contact centers, healthcare, finance, and government agencies routinely deploy IVR and AI routing; displacement is already widespread and accelerating. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Automated call routing, IVR, and AI receptionist services have been widely adopted across many industries for years, given clear ROI and mature vendor ecosystems. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI assists human operators by suggesting routes, auto-classifying messages, and summarizing caller intent, raising their throughput. However, the core task (routing) is so automatable that augmentation is less relevant than outright replacement in most contexts. |
| Augmentation potential | claude-sonnet-5 | 3/5 | For remaining human operators, AI transcription and message-summarization tools can speed up manual relay tasks, though the task itself is often fully automated rather than merely augmented. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Message routing and relaying is highly automatable via chatbots, IVR systems, and AI agents that can understand intent, classify messages, and direct them to appropriate recipients. Current AI systems can handle this end-to-end with substantial time savings, though some complex or ambiguous messages may still require human intervention. |
| Task automatability | claude-sonnet-5 | 5/5 | Message relaying and routing is a well-defined, text/voice-based task that current NLP and voice-transcription systems handle end-to-end with substantial time savings, e.g., automated attendants and message-routing bots.ID |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few legal or regulatory barriers exist to automating message routing itself; organizations do require some oversight of caller experience and occasional escalation protocols, but nothing mandates human handling. Customer preference for human contact is a mild friction point, not a hard barrier. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing or legal requirement for a human to relay or route messages; this task has long been substitutable for many contexts, though some customer preference for human contact may exist in high-touch answering services. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-based message routing (cloud-hosted chatbots, IVR) costs pennies per interaction versus human switchboard operator labor at $15–30/hour loaded. The cost differential is at least an order of magnitude in favor of automation. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated voice/message routing systems cost a small fraction of a human operator's wage per message handled, especially at scale via cloud telephony and NLP APIs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed products (IVR systems, chatbots, contact center automation platforms) reliably perform message routing and relaying in production at scale across thousands of organizations today. This is a mature, widely-implemented capability. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | IVR systems, chatbots, and AI-based virtual receptionists (e.g., automated answering services) are deployed at scale in production today, though some edge cases (garbled speech, ambiguous routing) still require human fallback. |
Place telephone calls or arrange conference calls as instructed.
88CI 79–97 · exposure 87 · augmentation 50 · importance 4.2/5 · click for rater detail
Place telephone calls or arrange conference calls as instructed.
88| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Enterprise communications and contact centers have been steadily automating call placement and conference scheduling for years. Modern organizations widely deploy IVR, call queuing, and automated scheduling systems in production, reflecting rapid and deep adoption in information and service sectors. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Telecommunications and office administrative functions have widely adopted automated calling and conferencing systems, though answering-service niches lag slightly behind faster-digitizing sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI assists operators by auto-suggesting call routing, managing hold queues, and scheduling conferences with minimal operator input. These tools raise productivity on routine calls but operators still manage complex transfers and problem resolution. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can assist by automating routine call setup, but the task itself is largely automatable rather than requiring augmented human judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Placing calls and arranging conferences is largely automatable using existing telephony APIs, VoIP systems, and call scheduling tools. The core task—dialing, connecting, and managing call routing—requires minimal human judgment and can achieve >50% time savings with current systems, though some edge cases (handling difficult routing, managing complex multi-party coordination) may require oversight. |
| Task automatability | claude-sonnet-5 | 5/5 | Placing calls or arranging conference calls is a simple, well-defined action easily handled by automated dialing systems, VoIP platforms, and AI voice assistants with full time savings at equal or better quality. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or liability barriers exist for basic call routing and scheduling; however, some organizations may require human oversight for certain call types, customer preference for human-initiated contact, and integration friction with legacy phone systems create modest friction. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There are no licensing, regulatory, or human-contact requirements that mandate a human perform simple call placement or conference setup. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | API-driven call placement and scheduling cost fractions of a cent per call versus the loaded wage of a switchboard operator (~$30–40/hour). Inference and integration overhead is negligible, making AI-driven automation orders of magnitude cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated dialing and conference setup costs fractions of a cent per call compared to paying a human operator's wage for the same task. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products reliably perform automated call placement and conference scheduling at scale (e.g., VoIP platforms, Twilio-based systems, calendar integrations). Production systems in many organizations already automate this task; minor limitations exist around natural conversation and handling unexpected call routing scenarios. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | Mature products (PBX systems, conferencing platforms like Zoom/Teams, virtual assistants like Alexa/Siri, and automated scheduling bots) already perform this reliably at scale in production today. |
Answer simple questions about clients' businesses, using reference files.
88CI 79–97 · exposure 87 · augmentation 63 · importance 4.2/5 · click for rater detail
Answer simple questions about clients' businesses, using reference files.
88| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Contact centers and service industries have rapidly deployed chatbots and IVR systems for handling routine inquiries; automation of simple FAQ-type calls is already widespread in telecommunications, healthcare, and professional services. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Answering services and call centers are rapidly adopting AI voice agents and chatbots, though full-scale replacement across all switchboard/answering service providers is still uneven. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist operators by pre-populating answers and reference suggestions as they talk to callers, reducing lookup time and improving consistency, though the task itself does not inherently require human judgment that augmentation would unlock. |
| Augmentation potential | claude-sonnet-5 | 4/5 | Even where humans remain involved, AI reference lookup and suggested-response tools significantly speed up answering routine client questions. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can reliably answer simple factual questions about businesses using knowledge bases and reference documents; this matches the 50%-time-saving threshold. A chatbot or agent can retrieve and synthesize basic business information (hours, services, contact details, policies) faster than a human operator without quality loss. |
| Task automatability | claude-sonnet-5 | 5/5 | Answering simple, reference-based questions about a business is a well-bounded retrieval and Q&A task that current conversational AI systems handle end-to-end with major time savings and equal or better consistency. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or legal barriers exist for automating simple information answers; however, some organizations prefer human contact for customer relationship and brand reasons, and some callers may demand a human operator. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, liability, or regulatory requirement mandates a human answer simple informational questions about a business. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference cost for answering a simple question is negligible ($0.001–0.01 per call), while a switchboard operator's fully loaded wage ($25–35/hour) means even a 2-minute call costs several dollars; AI is one to two orders of magnitude cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated voice/chat agents cost a small fraction per interaction compared to a human operator's loaded wage, especially at volume. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Multiple deployed products (IVR systems, chatbots, contact center AI) already handle simple business FAQs in production. Error rates are low for straightforward factual queries, though complex or edge-case questions may still require escalation. |
| Technical feasibility today | claude-sonnet-5 | 5/5 | AI-powered virtual receptionists and answering services (e.g., Ruby, Smith.ai's AI tiers, IVR/voicebot platforms) already deploy this in production at scale for many businesses today. |
Perform various data entry or word processing tasks, such as updating phone directories, typing or proofreading documents, or creating schedules.
84CI 76–92 · exposure 83 · augmentation 88 · importance 4.0/5 · click for rater detail
Perform various data entry or word processing tasks, such as updating phone directories, typing or proofreading documents, or creating schedules.
84| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Office automation and data entry have been among the fastest-adopting use cases in the information sector over the past 5 years. RPA and AI-powered document processing are in active production use across many large organizations. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Clerical and administrative support functions are adopting AI tools steadily but unevenly, with many smaller answering services and switchboard operations still relying on manual processes. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI significantly assists humans in these tasks—autocomplete, real-time proofreading, template-based scheduling suggestions, and intelligent form population all raise operator productivity while keeping human oversight in the loop. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI writing assistants, spell-checkers, and scheduling tools substantially boost productivity for these tasks while a human easily remains in the loop for final review. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Data entry, typing, proofreading, and schedule creation are routine, rule-based tasks that current AI systems (including text generation and form-filling agents) can perform end-to-end with significant time savings. Phone directory updates and document proofreading are well within the capabilities of modern LLMs and RPA tools. |
| Task automatability | claude-sonnet-5 | 4/5 | Data entry, proofreading, directory updates, and scheduling are highly structured text-manipulation tasks well within current LLM and automation tool capabilities, easily meeting the 50% time-saving bar. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | These are routine office tasks with minimal regulatory or legal barriers. The main friction is organizational inertia and desire to retain human oversight for accuracy, but nothing legally prevents full automation. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing, regulatory, or liability barriers restrict automating directory updates, typing, or scheduling; these are routine clerical functions with no legal requirement for human performance. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated data entry and proofreading via cloud APIs or RPA platforms cost a small fraction of an hourly wage, easily an order of magnitude cheaper when amortized across volume and factoring in no employee overhead. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | AI-based proofreading and data entry tools cost a small fraction of a cent per task compared to hourly wages for manual clerical work, representing an order-of-magnitude or greater cost advantage. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products including Microsoft Office automation, form-filling agents, and AI-powered proofreading tools (Grammarly, etc.) reliably perform these tasks in production. Some domain-specific nuances in scheduling or directory formats may introduce minor errors, but maturity is high. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Word processing software with AI-assisted proofreading, autocomplete, and data entry automation (e.g., RPA tools, spreadsheet/database sync) is deployed widely in production today, though directory-specific integrations vary by organization. |
Complete forms for sales orders.
80CI 67–92 · exposure 78 · augmentation 75 · importance 4.4/5 · click for rater detail
Complete forms for sales orders.
80| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Retail, e-commerce, and customer service sectors—where switchboard operators and order-taking occur—are adopting form automation and chatbot-integrated order processing rapidly. Production deployments are common in information and commerce-driven industries. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Answering services and call centers are adopting AI call handling and CRM automation at a moderate pace, with many pilots and growing production use but not yet universal deep adoption. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistants can suggest completions, validate fields in real-time, and flag errors before submission, significantly raising human operator productivity even when the human remains the primary order taker. This augmentation is commonly deployed today. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can pre-fill forms from transcribed calls or structured prompts, letting operators verify and correct rather than manually typing every field, meaningfully speeding the task while a human stays in the loop. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Completing forms for sales orders is highly structured data entry with a clear schema. Current AI systems can extract information from voice, email, or chat, validate against templates, and populate forms end-to-end—achieving well over 50% time savings compared to manual entry at equal accuracy. |
| Task automatability | claude-sonnet-5 | 4/5 | Filling out structured sales order forms from call information is a data-extraction and entry task well within the capability of modern AI voice/form systems with moderate setup, meeting the 50% time-saving bar in most cases. |
| Adoption barriers | claude-haiku-4-5-20251001 | 1/5 | Sales order form completion has no licensing requirement, legal mandate for human involvement, or significant liability asymmetry. Organizational adoption is largely a matter of technical integration and process confidence, not regulatory or contractual barriers. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing or legal requirement mandates a human complete a sales order form; the main friction is organizational inertia and ensuring data accuracy, not regulatory or liability barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Automated form completion via APIs and AI data extraction costs pennies per order, while a human switchboard operator handling this task costs $15–25 per hour fully loaded. The cost ratio is well over an order of magnitude in AI's favor. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated transcription and form-population software costs a fraction of a human operator's wage per transaction, though integration and error-checking overhead reduce the full order-of-magnitude savings somewhat. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature products (order management systems with AI data-entry capabilities, form-filling bots) reliably handle sales order form completion in production environments. Minor gaps exist in handling edge cases or custom field logic, but the core task is deployable at scale today. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | CRM and order-entry automation with voice-to-text and AI form-filling exists in production (e.g., IVR-to-CRM integrations), but many switchboard/answering-service contexts still rely on manual entry with only partial automation deployed. |
Answer incoming calls, greeting callers, providing information, transferring calls or taking messages as necessary.
79CI 79–79 · exposure 75 · augmentation 63 · importance 4.7/5 · click for rater detail
Answer incoming calls, greeting callers, providing information, transferring calls or taking messages as necessary.
79| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Adoption of AI call systems is already widespread in corporate and service sectors; thousands of organizations have replaced or are replacing human switchboard operators with automated systems. Measured displacement is visible across telecommunications, customer service, and administrative support. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | AI-based call answering and virtual receptionist services are being adopted rapidly across small business, healthcare, and customer service sectors, though full-scale replacement of dedicated operators is still uneven. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist remaining human operators by providing real-time transcription, caller-intent classification, and suggested transfers, raising efficiency. However, the task itself is increasingly automated rather than augmented, limiting the assistant role. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can pre-screen calls, transcribe messages, and draft routing/summaries, meaningfully boosting operator throughput and reducing routine handling time. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems can reliably answer calls, recognize speech, route to appropriate departments or voicemail, and log messages with high accuracy. While some complex caller scenarios may need human intervention, ~70-80% of routine calls can be handled fully automatically, meeting the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Modern conversational AI voice agents can greet callers, answer routine queries, route calls, and take messages with significant time savings, though edge cases and complex routing still need human fallback. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or licensure barriers exist for automated call-answering; customer preference for human interaction and organizational inertia are the main frictions. Some industries (healthcare, legal) may face liability concerns, but no hard legal requirement mandates human switchboard operators. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for this role, but some customer preference for human contact and organizational inertia in switching systems creates mild friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI call-handling per-call cost (infrastructure + inference + minimal oversight) is typically $0.01–0.10 per call, whereas a human switchboard operator costs $15–25/hour, translating to ~$1–3 per call accounting for handling time. AI is at least 10–100x cheaper. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated answering/routing services cost a small fraction of a human operator's wage per call handled, especially at volume. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Commercial IVR and AI call-handling systems (e.g., Google Voice AI, Vonage, Five9) demonstrably handle incoming calls, greetings, and routing in production at scale across many organizations. Error rates on simple greeting and transfer tasks are low, though occasional failures occur with accents or complex requests. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | AI call-handling and virtual receptionist products (e.g., IVR-plus-LLM systems, AI answering services) are deployed in production today at scale for many businesses, though accuracy on complex or ambiguous requests remains imperfect. |
Perform administrative tasks, such as accepting orders, scheduling appointments or meeting rooms, or sending and receiving faxes.
77CI 71–84 · exposure 70 · augmentation 63 · importance 3.7/5 · click for rater detail
Perform administrative tasks, such as accepting orders, scheduling appointments or meeting rooms, or sending and receiving faxes.
77| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Information and professional services sectors have widely deployed automated scheduling, order systems, and contact center automation; adoption is rapid and measurable, though slower in small firms and legacy organizations. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Adoption of AI scheduling and virtual receptionist tools is growing but still uneven, with many small businesses and answering services relying on human operators. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI assists humans by pre-screening orders, suggesting appointment times, and organizing fax queues, raising operator efficiency on the portions requiring human judgment or specialized handling. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools significantly assist operators by pre-filling forms, suggesting scheduling slots, and drafting responses, improving throughput while humans oversee exceptions. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | The core administrative tasks (accepting orders, scheduling, fax handling) are highly routine and well-structured, amenable to automation via current AI/RPA systems. End-to-end automation with significant time savings is achievable, though integration complexity and occasional edge cases prevent a perfect 5. |
| Task automatability | claude-sonnet-5 | 4/5 | AI voice assistants and scheduling bots can already handle order intake, appointment booking, and routing with significant time savings, though edge cases still require human handling. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or legal barriers exist for automating these administrative tasks; most friction is organizational (customer preference for human touch, internal workflow integration). No licensing requirement mandates human involvement. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or legal requirement mandates a human perform basic administrative scheduling or order intake tasks. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-based order systems, scheduling bots, and fax-to-email automation cost a fraction of a human switchboard operator's loaded wage, easily achieving an order-of-magnitude cost advantage. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated voice/chat systems for scheduling and order-taking cost a small fraction of a human operator's wage per interaction at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products (RPA platforms, chatbots, calendar APIs, email automation) reliably handle order processing, appointment scheduling, and fax routing in production environments. Material error rates exist for ambiguous requests, keeping this below a 5. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Conversational AI scheduling and order-taking products (e.g., AI receptionists, calendar bots) are deployed in production but still have noticeable error rates and narrower scope than a full human operator. |
Place orders, such as for equipment, supplies, or catering for meetings.
76CI 67–84 · exposure 70 · augmentation 75 · importance 3.4/5 · click for rater detail
Place orders, such as for equipment, supplies, or catering for meetings.
76| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Enterprise procurement and administrative sectors have adopted automation for order placement at significant scale; many mid-to-large organizations use e-procurement or chatbot systems in production. Smaller and less digitized sectors lag, but overall velocity in information and service sectors is strong. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Office/administrative support functions are adopting AI tools at a moderate pace, with procurement automation growing but not yet universal, especially in smaller organizations relying on switchboard/answering services. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistants can draft orders, suggest vendors, auto-populate details from prior orders, and flag compliance or budget issues, substantially reducing the time and cognitive load for a human operator managing order requests. The human retains final review and context judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can significantly speed up drafting orders, checking inventories, and generating catering requests, letting the human operator focus on exceptions and communication. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Current AI systems can reliably process structured order information (item, quantity, vendor, delivery details) and submit orders through APIs or web forms with minimal human intervention, achieving the 50% time-saving threshold. However, some contextual judgment about specifications, vendor selection, or cost optimization may still benefit from human review. |
| Task automatability | claude-sonnet-5 | 4/5 | Placing routine orders for equipment, supplies, or catering is a structured, rules-based task that AI agents integrated with procurement or ordering systems can largely complete, including selecting vendors, filling forms, and confirming orders.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few hard legal or licensing barriers exist for automating routine order placement; main friction points are internal organizational preferences for human touch and approval workflows that may require management sign-off rather than legal mandate. |
| Adoption barriers | claude-sonnet-5 | 1/5 | There is no licensing or legal requirement for a human to place equipment or catering orders; it's a purely administrative task with minimal regulatory barriers. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven order placement (via integration with procurement systems or agents) costs a small fraction per transaction compared to a human operator's loaded wage; at scale, the cost differential is an order of magnitude in favor of automation. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated ordering via software/AI agents costs a small fraction of a human operator's time-based wage for repetitive transactional ordering tasks. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products including e-procurement platforms, chatbots, and RPA solutions routinely handle order placement tasks in production environments across enterprises. Error rates are low for standardized orders, though edge cases and vendor-specific requirements occasionally require human oversight. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Procurement automation and AI ordering assistants exist and are used in some enterprises, but many organizations still rely on human coordination for exceptions, approvals, or vendor negotiation, so reliability varies. |
Monitor alarm systems to ensure that secure conditions are maintained.
71CI 59–84 · exposure 67 · augmentation 63 · importance 4.6/5 · click for rater detail
Monitor alarm systems to ensure that secure conditions are maintained.
71| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Security and facility management sectors are actively adopting AI-powered monitoring systems in production. Major enterprises, financial institutions, and data centers have already shifted to automated alerting; adoption is accelerating as legacy systems are replaced and ROI is proven. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Security and monitoring industries have adopted automated systems steadily, but many answering/monitoring services remain small firms with slower, partial AI integration. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI provides meaningful assistance by filtering false alarms, prioritizing events by severity, and providing real-time pattern analysis; human operators can focus on high-confidence threats and nuanced decision-making. However, the task itself lends itself more to replacement than augmentation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI-assisted alarm triage and pattern recognition significantly help human operators prioritize and respond faster while retaining human decision-making for critical judgment calls. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Alarm system monitoring is inherently automatable—AI systems today can ingest sensor data, detect anomalies, classify threats, and trigger alerts with high reliability. The task requires minimal human judgment and is primarily pattern-matching and alerting, which modern AI excels at; the main manual overhead is investigation and escalation, which can be partially automated. |
| Task automatability | claude-sonnet-5 | 3/5 | AI-based monitoring systems can detect and flag alarm events automatically, but full end-to-end handling including verification, escalation, and human judgment calls on ambiguous alerts still requires human oversight, limiting the full time-saving threshold. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While some facilities may have contractual or regulatory requirements for human oversight (e.g., physical security contracts, compliance audits), the core monitoring task itself has no hard legal barrier—AI systems are widely accepted and often legally sufficient. Customer preference for human operators provides mild friction but is eroding. |
| Adoption barriers | claude-sonnet-5 | 2/5 | There are modest liability concerns (false alarms, missed emergencies) and some contractual/customer expectations for human responsiveness, but no licensing requirement mandates a human specifically monitor alarms. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Cloud-based AI monitoring services and SIEM platforms cost a fraction of human operator wages when amortized per facility or per alarm event monitored. A single AI system can monitor thousands of sensors simultaneously, achieving orders-of-magnitude cost advantage over human staffing. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated monitoring software scales cheaply across many alarm points compared to paying human switchboard operators per shift, though some human oversight cost remains. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Deployed security monitoring systems, SIEM (Security Information and Event Management) platforms, and AI-powered anomaly detection are mature production systems used at scale in enterprises, data centers, and critical infrastructure. These systems demonstrably perform reliable 24/7 monitoring and alerting today. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated alarm monitoring and dispatch products exist and are deployed in security/monitoring centers, but many still route ambiguous or high-stakes alerts to human operators due to error-rate concerns. |
Record messages, suggesting rewording for clarity or conciseness.
71CI 59–84 · exposure 62 · augmentation 75 · importance 4.5/5 · click for rater detail
Record messages, suggesting rewording for clarity or conciseness.
71| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Call centers, customer service, and administrative sectors are already adopting speech-to-text and automated message handling at significant scale. Cloud-based answering services and voicemail-to-text solutions are mainstream. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Answering services and reception functions are adopting AI at a moderate pace, with many small businesses still using human operators or hybrid setups, but growing rapidly. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assists switchboard operators by auto-generating transcripts and draft rewording suggestions in real-time, significantly reducing manual editing and transcription time while the operator retains control over message quality and final form. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools substantially help operators by drafting clearer versions of messages and flagging ambiguities, improving speed and quality when a human remains in the loop. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | AI can transcribe messages and suggest edits for clarity/conciseness with moderate accuracy, but requires human judgment on context-specific appropriateness and often needs review. The task involves both mechanical transcription (automatable) and nuanced language improvement (partially automatable). |
| Task automatability | claude-sonnet-5 | 4/5 | Modern speech-to-text and LLM systems can transcribe calls, summarize messages, and suggest clearer/more concise wording largely automatically, meeting the time-saving threshold for most routine messages.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Few regulatory or legal barriers exist for automating message recording and suggestion. Organizational and customer-preference friction exists (some callers expect human operators), but no licensing or hard legal requirement prevents AI substitution. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or legal requirement mandates a human perform message-taking and rewording; it's a low-liability clerical task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI transcription and editing tools cost pennies per message when amortized across volume, substantially cheaper than paying a human operator to listen, transcribe, and suggest edits. Integration overhead is modest for modern organizations. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | AI transcription and summarization costs fractions of a cent per message versus a human operator's wage, making it drastically cheaper at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Speech-to-text and text-refinement tools exist in production (e.g., Otter.ai, Descript), but error rates on accents, technical jargon, and domain-specific terminology remain material. These systems work adequately for simple messages but falter on complex or contextually sensitive rewording suggestions. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Deployed virtual receptionist and call-answering AI products (e.g., AI answering services, voicemail transcription with summarization) already do this in production, though accuracy on noisy audio or ambiguous requests can require review. |
Process incoming or outgoing mail, packages, or deliveries.
69CI 56–82 · exposure 70 · augmentation 50 · importance 4.0/5 · click for rater detail
Process incoming or outgoing mail, packages, or deliveries.
69| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Postal services, logistics companies, and large enterprises have deeply adopted mail and package automation at scale for decades; adoption is rapid and pervasive in information and logistics sectors, with continuous investment in automation. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | This task sits within low-digitization, physical-operations contexts (mailrooms, answering services) where automation adoption has been slower than in pure information-processing sectors. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI and automation assist human operators by providing real-time tracking, automated sorting flags, and delivery status alerts, raising their efficiency on exception handling and customer inquiries. Humans remain useful for problem-solving and customer service aspects. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-driven tracking, sorting software, and automated notifications help operators manage and prioritize mail/package flow, improving efficiency without replacing physical handling steps. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | Most of this task is highly automatable: mail sorting can be done by machine vision and robotic systems, package tracking is fully automated, and delivery routing is handled by logistics algorithms. The remaining human element (signing for deliveries, handling exceptions) represents <50% of typical effort, meeting the automatability threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Sorting, routing, and logging mail/packages is largely rule-based and physical-scanning work that AI-enabled systems (barcode/label recognition, routing software) can handle for most digital sub-tasks, though physical handling still requires human or robotic intervention. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Moderate barriers exist: some regulatory requirements for signature capture and proof of delivery, customer preferences for human contact on exceptions, and organizational inertia in smaller answering services and offices. However, no legal requirement mandates a human perform mail processing itself. |
| Adoption barriers | claude-sonnet-5 | 1/5 | No licensing or legal requirement mandates a human to process mail or packages; it's a low-liability, non-regulated clerical/physical task. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Automated mail and package handling systems cost significantly less per unit processed than a human operator; modern postal and logistics automation achieves sub-dollar cost per item, while human labor loaded cost is much higher per transaction processed. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated sorting/tracking software is cheap to run, but the physical handling, exception management, and integration costs keep overall cost roughly comparable to human labor in smaller operations. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed systems reliably handle mail/package sorting (postal service automation), tracking (USPS, UPS, FedEx platforms), and routing (logistics software). Physical robot arms and conveyor systems for mail processing are in production at scale in large mail facilities, though last-mile human handling persists. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated mailroom and package-tracking systems exist and are used in many organizations, but full end-to-end automation (physical sorting, delivery confirmation, exception handling) still commonly relies on human staff. |
Operate communication systems, such as telephone, switchboard, intercom, two-way radio, or public address.
54CI 30–79 · exposure 50 · augmentation 50 · importance 4.7/5 · click for rater detail
Operate communication systems, such as telephone, switchboard, intercom, two-way radio, or public address.
54| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption has been slow and inconsistent; while large enterprises use IVR for call filtering, many mid-market and small organizations still rely on human switchboard operators, and adoption of full automation remains niche despite decades of technology availability. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Call centers and answering services have rapidly adopted AI-driven IVR and virtual receptionist tools, though full replacement is uneven across smaller firms. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist operators via call transcription, automatic call logging, and intelligent routing suggestions, raising productivity modestly. However, the operator remains central to handling unpredictable edge cases and complex transfers, so augmentation impact is partial rather than transformative. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI can pre-screen, transcribe, and route calls to assist remaining human operators, improving throughput on routine communication management tasks. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While call routing logic can be automated via IVR and call-answering systems, the full task—operating multiple communication systems, handling operator discretion, managing complex switchboard features, and providing human judgment—remains largely manual. Current AI handles narrow subsets (voicemail transcription, basic routing) but not end-to-end operation at 50% time saving. |
| Task automatability | claude-sonnet-5 | 4/5 | Modern AI voice agents and cloud PBX/IVR systems can route calls, answer, transfer, and manage intercom-like functions with substantial time savings, though edge cases (emergencies, complex routing) still need human oversight. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Moderate barriers exist: many organizations require human operators for emergency lines, customer preference for human contact on sensitive calls, and integration friction with existing proprietary telephone systems. Regulatory requirements (e.g., emergency 911 compliance) add some constraint, though not a hard legal mandate that a human must operate. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement for this role; main friction is customer preference for a human voice and quality/liability concerns in sensitive contexts like emergency lines. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | IVR and auto-attendant infrastructure is non-trivial to implement and maintain; integration with legacy switchboards and oversight costs remain significant relative to a low-wage switchboard operator in many markets. |
| Cost vs. human wage | claude-sonnet-5 | 5/5 | Automated switchboard/answering software costs a small fraction of a human operator's wage per call handled, especially at scale. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | IVR systems and auto-attendants exist in production but typically handle only simple call routing; they fail on complex queries, transfers, and real-time troubleshooting. No deployed product reliably replaces a switchboard operator across the full range of communication systems and contexts. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Deployed virtual receptionist and call-routing products (e.g., cloud answering services, AI IVR) already handle high call volumes reliably in production for many businesses today. |
Greet visitors, log them in and out of the facility, assign them security badges, and contact employee escorts.
50CI 25–75 · exposure 50 · augmentation 50 · importance 4.7/5 · click for rater detail
Greet visitors, log them in and out of the facility, assign them security badges, and contact employee escorts.
50| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Adoption is slow and patchy, concentrated in large enterprises or high-security environments with capital investment; small and mid-size firms, call centers, and service facilities still rely on human operators. Measured displacement remains minimal sector-wide. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Corporate offices, especially in tech, finance, and professional services, have rapidly adopted digital visitor management and self-service kiosks over the past decade. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | Digital check-in systems and automated escort notifications can assist operators by reducing manual data entry and speeding communication, but the core task—human greeting, identity verification, and badge issuance—remains operator-dependent and sees only moderate productivity lift. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI assists switchboard/reception staff by automating routine logging and badge issuance, freeing them to handle exceptions, but still requires oversight for security-sensitive edge cases. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | Greeting and logging visitors involves some routine data entry that could be partially automated (kiosks, digital check-in), but assigning physical security badges and real-time contact coordination with employee escorts requires human judgment and physical interaction. Current systems can handle 20–30% of the process reliably, falling short of the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Visitor check-in, logging, badge assignment, and escort notification are structured, rule-based tasks that AI-driven visitor management kiosks and chatbots already handle largely autonomously, with human intervention only for exceptions. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Physical security, liability, and facility access control create significant friction; many organizations legally require a human to verify identity, assign credentials, and maintain a live security presence for compliance and emergency response. Customer preference and security regulations strongly favor human contact. |
| Adoption barriers | claude-sonnet-5 | 2/5 | Some facilities require human judgment for security exceptions or physical badge handling, but there is no licensing or legal requirement that a human perform this greeting/logging function. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Partial automation (kiosk + badge dispenser + alert system) can reduce labor cost by 30–40%, but capital setup, maintenance, and human oversight for edge cases mean the all-in cost remains comparable to or slightly cheaper than a single operator—not an order of magnitude difference. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Kiosk/software subscriptions cost a small fraction of a receptionist's wage for high-volume, repetitive check-in tasks, though some hardware/setup costs exist. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Self-service kiosks and automated visitor management software exist in some facilities, but they operate in controlled, coordinated environments and still rely on human badge assignment and escort notification. No end-to-end deployed system reliably handles all elements (greeting, logging, badging, escort contact) without human intervention. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Deployed visitor management systems (e.g., Envoy, SwipedOn) reliably automate sign-in, badge printing, and escort alerts in production across many corporate facilities today. |
Monitor emergency and code alarms, make emergency announcements, or route emergency calls to the appropriate location.
34CI 25–43 · exposure 38 · augmentation 50 · importance 4.5/5 · click for rater detail
Monitor emergency and code alarms, make emergency announcements, or route emergency calls to the appropriate location.
34| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Switchboard and answering services remain concentrated in small to mid-sized firms, healthcare, and government—sectors with slower digitization and high regulatory caution around emergency automation. Actual deployment of AI-only emergency routing is minimal; most adoption remains at the pilot or supplemental level. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Switchboard/answering service work sits in a low-digitization, often small-business or facility-services context, with automation focused narrowly on alarm systems rather than broad AI-driven operational deployment. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist operators by flagging call urgency, suggesting routing, transcribing caller information, and logging incidents, meaningfully improving operator efficiency and accuracy. However, the human remains the critical decision-maker, making this augmentation useful but not transformative. |
| Augmentation potential | claude-sonnet-5 | 3/5 | AI-assisted alarm monitoring, alerting, and call-routing tools can help operators triage and respond faster, but human judgment remains central for emergency communication and coordination. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can detect alarm signals and route calls based on simple logic, emergency response requires nuanced judgment about urgency, caller distress assessment, and dynamic rerouting based on context—capabilities current systems handle inconsistently. End-to-end automation falls short of the 50% time-saving threshold because human verification remains essential for safety-critical decisions. |
| Task automatability | claude-sonnet-5 | 3/5 | AI voice/IVR systems can detect keywords and route many routine calls, but emergency triage requires judgment about ambiguous, high-stakes situations that still typically need human confirmation, limiting full end-to-end automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Emergency call handling faces substantial regulatory barriers: many jurisdictions mandate human operators or legal accountability for emergency routing; liability exposure for misrouted or delayed emergency calls is severe; and organizations face organizational and reputational friction around replacing humans in life-safety contexts. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Emergency response often carries regulatory, safety-code, and liability requirements that push facilities to keep a human as final decision-maker or fallback, creating strong barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI infrastructure (cloud telephony, speech recognition, routing engines) incurs significant ongoing costs for integration, monitoring, and fail-safe mechanisms required in emergency contexts. The all-in cost per call remains comparable to or higher than a trained operator's loaded wage when liability and redundancy requirements are factored in. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated monitoring/routing software has low marginal cost, but integration with facility-specific emergency protocols and required human backup oversight keeps blended costs closer to parity rather than order-of-magnitude cheaper. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Some AI-assisted call routing exists in production (IVR systems), but reliable emergency detection, accurate triage, and proper escalation under high-stakes conditions remain largely manual or hybrid. No deployed product reliably handles the full spectrum of emergency scenarios without human oversight. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Automated call routing and alarm monitoring systems are deployed in many facilities, but emergency-specific announcement and routing systems typically retain human oversight due to error consequences, so reliability at scale for the full task is not yet universal. |
Perform various cash handling tasks, such as collecting payments, making bank deposits, or managing petty cash.
21CI 18–25 · exposure 20 · augmentation 38 · importance 4.2/5 · click for rater detail
Perform various cash handling tasks, such as collecting payments, making bank deposits, or managing petty cash.
21| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | Switchboard operator positions are concentrated in small firms, call centers, and administrative roles with lower digitization and slower technology adoption. While digital payment systems are spreading, the specific task of cash handling by switchboard operators remains largely manual and non-automated in these sectors. |
| Sector adoption velocity | claude-sonnet-5 | 1/5 | Switchboard/answering service roles are low-digitization, and cash handling specifically has seen minimal AI-driven displacement in practice. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI-powered accounting and payment software can assist operators by automating ledger entry, flagging discrepancies, and generating deposit summaries, raising productivity on the administrative portions of cash handling. However, the augmentation is limited to portions of the task; physical cash verification and reconciliation remain largely human-dependent. |
| Augmentation potential | claude-sonnet-5 | 2/5 | Accounting software or apps can help track petty cash and reconcile records, but the core physical handling and deposit tasks receive little AI assistance. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While payment collection and deposit preparation involve structured, repetitive data entry that AI could partially automate (e.g., processing digital payments, categorizing transactions), the physical handling of cash, verification of authenticity, and reconciliation of discrepancies require human oversight. End-to-end automation with 50% time savings at equal quality is not achievable today without significant human involvement in verification and exception handling. |
| Task automatability | claude-sonnet-5 | 2/5 | Cash handling involves physical collection, counting, and bank deposit trips that current AI cannot perform end-to-end; only record-keeping portions could be aided by software. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Strong regulatory and liability barriers exist: financial institutions, payment processors, and employers face strict compliance requirements (SOX, anti-fraud regulations, audit trails) that legally require human accountability for cash handling. Custody and reconciliation responsibilities create legal and fiduciary obligations that cannot be fully transferred to AI systems. |
| Adoption barriers | claude-sonnet-5 | 4/5 | Handling money involves accountability, fraud risk, and often bonding/authorization requirements, plus physical presence needed for deposits and cash custody. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | Current AI solutions for cash handling (payment processors, accounting software) require significant human oversight and do not reduce the total cost below or substantially below a human switchboard operator's loaded wage when integration, verification, and liability costs are included. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Physical cash handling still requires a human physically present, so AI cannot substitute cheaply; at best software reduces bookkeeping overhead marginally. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | No deployed product reliably performs full cash handling autonomously in production. Partial automation exists for digital payment processing and accounting ledgers, but physical cash collection, counting, verification, and bank deposit procedures remain manual tasks with human accountability in real organizations. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | No deployed AI product physically handles cash, counts money, or makes bank deposits; this remains a physical-world task outside current AI product scope. |
Contact security staff members when necessary, using radio-telephones.
13CI 0–25 · exposure 8 · augmentation 25 · importance 4.6/5 · click for rater detail
Contact security staff members when necessary, using radio-telephones.
13| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 1/5 | Security operations remain labor-intensive and highly resistant to automation due to liability, regulatory scrutiny, and the need for human judgment. Switchboard operators for security have not been displaced by AI in production settings. |
| Sector adoption velocity | claude-sonnet-5 | 2/5 | Switchboard/answering service roles are a declining, low-digitization niche with little evidence of AI agents being deployed for physical security radio dispatch tasks. |
| Augmentation potential | claude-haiku-4-5-20251001 | 2/5 | AI could potentially assist by logging communications or prioritizing which security personnel to contact based on incident type, but the core act of initiating the radio contact and ensuring receipt remains fundamentally human-driven, limiting augmentation value. |
| Augmentation potential | claude-sonnet-5 | 2/5 | AI could potentially flag anomalies or assist in triaging when to alert security, but for the specific act of contacting staff via radio-telephone, current tools offer minimal direct assistance. |
| Task automatability | claude-haiku-4-5-20251001 | 1/5 | Contacting security staff via radio-telephone is a real-time, synchronous communication task that requires immediate human presence and judgment on the other end. Current AI cannot reliably initiate, conduct, or resolve such communications with the responsiveness and contextual nuance required, nor can it replace the human decision-making about when contact is 'necessary.' |
| Task automatability | claude-sonnet-5 | 2/5 | This is a brief, situational communication act requiring judgment about when and whom to contact, which is not a task AI can meaningfully replace end-to-end given the physical radio-telephone interface and real-time human decision-making involved. |
| Adoption barriers | claude-haiku-4-5-20251001 | 5/5 | Security communications are heavily regulated (NFPA, FCC rules on radio use, emergency services protocols), and liability for failed or incorrect emergency contact is severe. A human must typically be responsible for critical security notifications, creating a hard legal and organizational barrier to automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not licensed work, this task involves safety/security escalation where organizations typically want a human accountable for judgment calls and equipment operation, creating moderate organizational friction against automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 1/5 | The cost of deploying and maintaining specialized radio infrastructure, compliance systems, and oversight for AI-initiated security communications would exceed the cost of a human operator handling such calls, especially given the low volume and critical nature of security contacts. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Building and maintaining a reliable AI system to detect the need for security contact and execute radio-telephone communication would likely cost more than the marginal human effort for this narrow, occasional action. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed AI system can autonomously operate radio-telephone equipment to initiate emergency or routine contact with security staff in a way that would be reliable or legally acceptable in practice. This is not a solved task in production environments. |
| Technical feasibility today | claude-sonnet-5 | 1/5 | There are no deployed products that autonomously monitor for security-relevant events and independently contact security staff via radio-telephone in operator settings today. |
Related occupations — Office & Administrative Support
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.