More answers like this
GPT-5Sep 30, 2026openai/gpt-5-2025-08-07
Will AI create or take away more job opportunities?
Short answer: both—but in different ways and on different timelines. AI mostly automates tasks inside jobs before it replaces whole jobs, and it also creates new tasks, roles, and even industries. Whether the net effect is job-creating or job-reducing will depend on how quickly organizations redesign work and how much complementary demand and policy support show up.
What’s likely over the next 2–5 years
•High churn, not a collapse: many roles will change shape; some shrink, some grow. Task mixes will shift monthly, not just yearly.
•Biggest pressure: routine, text-heavy office work—customer support, basic coding, bookkeeping, ad ops, paralegal drafting, claims processing, data entry, templated research and reporting.
•Biggest lift: roles where judgment, relationships, creativity, or physical presence matter—healthcare, skilled trades, teaching/tutoring, sales, product and UX, robotics maintenance, field service.
•Multipliers: people who learn to direct AI (workflows, prompts, tooling) get more productive and scarce.
New and expanding job families
•AI product owners, AI workflow designers, “copilot” enablement for functions (sales, service, finance).
•Data curation/stewardship, evaluation, model risk/audit, compliance and safety.
•MLOps/AI platform engineering, synthetic data and simulation, edge/robotics techs.
•Short-lived niche titles (e.g., “prompt engineer”) will fold into broader roles.
Wages and inequality
•Complemented workers (who use AI to amplify expertise) tend to see wage gains.
•Mid-skill routine cognitive work sees the most pressure.
•Many in-person service jobs are less exposed to automation but can be augmented.
What individuals can do now
•Treat AI as a copilot: adopt one general model and one function-specific tool; measure time saved on 3–5 recurring tasks.
•Move up the value chain: do less drafting/formatting, more scoping, verifying, deciding, and persuading.
•Build data/automation literacy: basic Python or no-code automation, analytics, and prompt design for your domain.
•Sharpen “human moat” skills: domain expertise, communication, client handling, ethics, and negotiation.
•Create proof of work: a small portfolio of before/after AI-augmented projects with quantified impact.
•Hedge: cultivate T-shaped skills and a network across functions; consider a side project that uses AI to deliver a service.
What organizations and policymakers can do
•Redesign jobs around human-AI teaming; measure outcomes, not keystrokes.
•Retrain incumbents; give them time and incentives to learn tools.
•Create guardrails: accuracy checks, data governance, and model risk management.
•Encourage job creation with innovation, diffusion, and support for small firms adopting AI.
If you tell me your industry and role, I can map which of your tasks are most automatable, which are high-value complements, and suggest a 90-day upskilling plan with tools and practice projects.