More answers like this
GPT-5Oct 27, 2026openai/gpt-5-2025-08-07
Will AI create or take away more job opportunities?
Both sides
Short answer: both. AI will eliminate some tasks and roles, create new ones, and redesign many more. Whether it nets out to “more jobs” or “fewer” depends on how fast organizations adopt AI, how they reinvest productivity gains, how elastic demand is in each market, and whether workers can reskill quickly enough.
How it tends to play out
•AI replaces tasks before it replaces whole jobs. Many roles will see 10–50% of their work automated and be re-scoped rather than fully cut.
•Productivity gains can raise labor demand if cheaper, better products expand the market (think: more customer support available 24/7 increases sales, which funds more human roles).
•The transition brings churn: some occupations shrink even as new ones grow, with pressure on mid-wage, routine cognitive work and opportunities in roles that combine tech with domain expertise or human contact.
Roles most exposed to reduction or redesign
•Routine content creation and editing, basic customer support, data entry/transcription, bookkeeping, ad-ops, basic code scaffolding, first-pass legal/paralegal review, and some diagnostic triage. Entry-level knowledge-work ladders may narrow.
Roles likely to grow
•AI/ML engineering, data and analytics, prompt/tooling developers, MLOps, cybersecurity, product and growth roles that wield AI, human-in-the-loop QA, AI safety/governance/compliance, change management and training.
•Human-centric work: healthcare, education/tutoring, skilled trades and field techs, sales, account management, creative direction, project/program management.
•Entrepreneurship and solo businesses, as AI lowers the cost to build, market, and operate.
What this means for you
•Specialize, then augment: pick a domain and learn to use AI to 10x that specialty rather than trying to “be the AI.”
•Automate your own job first: build small scripts/agents to remove your drudgework; document the ROI.
•Lean into human moats: judgment, taste, leadership, negotiation, field work, and relationships.
•Get data- and tool-fluent: prompt engineering, basic Python/SQL, workflow automation (e.g., Zapier/Make), and model strengths/limits.
•Show outcomes, not effort: keep a portfolio of before/after metrics from AI-augmented projects.
•Keep learning loops tight: quarterly skills refresh, peer groups, and shipping small experiments.
If you share your field and level (student, early career, mid, switcher), I can map the highest-impact AI tools to learn, tasks to automate first, and a 60–90 day upskilling plan.