AI-Augmented Versions of Common Roles

The augmented version of a role is not a reduced version. It typically involves less volume work and more judgment, communication, and taste -- the things AI cannot reliably do. This module maps five common roles through that transition. McKinsey's 2023 work on AI and the future of work estimated that 60-70% of employee time across all occupations is currently spent on activities that could be partially automated; the World Economic Forum Future of Jobs Report 2023 estimated that 44% of worker skills will be disrupted over five years. Both ranges are aggregate. What follows is what those numbers actually mean at the desk level, role by role.

Accountant / Finance Analyst

Routine data extraction, reconciliation, transaction categorisation, variance commentary, and standard report formatting are high AI-exposure tasks. Tools like Excel's Copilot, Microsoft 365 Copilot, and finance-specific platforms now produce useable first drafts of monthly management accounts from structured ledger data. The augmented version of the role shifts time toward interpreting anomalies, advising on tax strategy for novel situations, communicating financial risk to non-finance stakeholders, and managing relationships with auditors and regulators. Technical accounting knowledge stays essential -- the AI produces outputs that a qualified professional still has to validate, and in regulated contexts the human review is non-negotiable.

The augmented accountant is increasingly an internal advisor, not a producer of reports. The reports get produced anyway; the value is what gets said about them in the room. That requires presenting financial information in plain language to non-finance colleagues and helping them make better operational decisions.

Marketing Manager

First-draft copy, A/B test variants, audience segmentation from structured data, campaign performance summaries, social-post variations, and ad-creative iterations are high exposure. Almost every modern marketing stack now has AI features generating exactly these outputs. The augmented role concentrates on brand judgement -- what is on-brand when an AI can produce plausible-sounding copy in any voice -- strategy (which channels and messages for which audience at which stage), and relationships with agencies, media partners, and sales teams.

The harder shift is in measurement. AI tools make it easy to run more experiments, generate more variants, and produce more reports. That increases the volume of data but not necessarily the volume of insight. The augmented marketer becomes better at choosing what not to measure, what experiments not to run, and which channel results to ignore as noise. Strategic restraint becomes more valuable as production becomes cheaper.

Writer / Content Strategist

Formulaic content -- product descriptions, FAQs, templated articles on known topics, transcript-based summaries -- has high AI exposure. The augmented writer concentrates on original research and reporting (going to the source, not paraphrasing existing content), distinctive voice and point of view, editorial judgement about what to cover and how, and structured persuasion that requires understanding a specific reader's situation. SEO-driven content that exists only to capture queries is increasingly produced and ranked by AI on both sides; that whole layer is collapsing toward zero margin.

A different shift in publishing: AI Overviews on Google and answer engines on ChatGPT, Perplexity, and Claude are reducing the click-through rate on commodity informational content. The writer who survives the next two years is one whose work is being cited as the source by those engines, not paraphrased into them. That requires either original reporting, named expertise, or a position the engine cannot derive from scraping existing content.

Customer Support Specialist

L1 ticket resolution, FAQ responses, status updates on known issues, and standard refund processing are high exposure. Modern customer-support AI now resolves measurable shares of L1 volume without escalation -- Klarna publicly reported in 2024 that its AI assistant handled work equivalent to roughly 700 full-time agents on volume that was previously L1. The augmented support specialist handles escalations that require empathy and de-escalation, complex product issues with no prior resolution path, internal coordination with engineering, and identifying systemic issues from patterns in customer feedback that no individual ticket would surface.

A second shift: the augmented support function becomes a primary product-feedback loop. Specialists who can synthesise hundreds of conversations into a clear "here is what customers are actually struggling with this month" become more valuable, because that synthesis is a judgement task that AI assists but does not replace. Routing that signal to product and engineering with credibility is the new value layer.

Designer

Generating visual variations, resizing assets for multiple formats, producing stock-style imagery, drafting wireframes from a brief, and producing variations of a known layout are high exposure. The augmented designer focuses on concept development and creative direction, understanding a brief well enough to evaluate AI output against it, brand consistency judgement, and communication with stakeholders who cannot evaluate AI output themselves. Senior designers are spending more time on prompt-and-curate workflows and less time on production.

A second shift specific to product designers: as AI lowers the cost of producing UI variations, the bottleneck moves to deciding which variation to ship based on user research, accessibility constraints, business logic, and engineering cost. Designers who pair tightly with PM and engineering on that decision become more central; designers whose value was production speed lose ground.

The common pattern across all five roles

Volume work and pattern work moves toward AI. Judgement, communication, and taste become the primary value. That does not make roles easier -- those skills are harder to develop, take longer to acquire, and are less commoditised. The skills worth investing in are different from the ones that got most of the professional development attention in the pre-AI era. The next module unpacks that skill stack in detail; the practice block below gets you started on the version specific to your role.