JobForesight's Task Index Shows Which AI Roles Actually Survive

JobForesight dropped their 2026 AI Career Risk Index this week with a counterintuitive finding: the variance inside your job matters more than the variance between jobs.

The open dataset, released under Creative Commons license, fills a gap that OECD, ILO, and Anthropic all acknowledge but haven't addressed. While Anthropic publishes observed Claude usage by task, nobody else offers per-occupation task-replaceability scores under open license.

The Task Revolution

"AI's impact happens at the task level, not the job level," the report states. This aligns with what hiring managers are discovering in practice: two software engineers with identical titles can face wildly different AI exposure based on their actual daily tasks.

The methodology, detailed by founder Robiul Islam, combines published research with editorial judgment. Islam takes an unusually transparent approach, explicitly stating the model's limitations rather than overselling any single figure.

For AI operators, this granular view changes everything. Instead of asking "Will AI replace software engineers?" the question becomes "Which engineering tasks remain human-centric?"

What the Numbers Reveal

The index shows surprising resilience in roles traditionally considered vulnerable. Senior positions that involve complex decision-making, stakeholder management, and creative problem-solving show lower task-level exposure than their junior counterparts.

This tracks with recent DORA data from Lushbinary showing AI raised pull request output 98% but increased incidents 242%. DX Research found median productivity gains of only 7.76% despite 65% more AI usage. The pattern: AI excels at volume but struggles with judgment calls.

SkillPanel's workforce exposure metrics confirm this trend. Organizations using task-level analysis report more accurate workforce planning than those relying on job-title assessments alone.

Practical Applications for Hiring

Smart hiring managers are already adapting. Instead of filtering by job titles, they're mapping candidate skills against task-exposure scores. A backend engineer who spends 80% of their time on boilerplate code faces different prospects than one architecting distributed systems.

The AIJobImpactCalculator.com, which cross-references multiple sources including JobForesight's data, shows 78 million net new roles emerging by 2030 according to the WEF Future of Jobs Report. That's 170 million created against 92 million displaced. The key: knowing which tasks within those roles actually matter.

For talent acquisition teams, this means rewriting job descriptions around irreplaceable tasks. For candidates, it means highlighting work that demonstrates judgment, creativity, and complex reasoning.

Strategic Implications

The task-level view also explains why some AI transformations fail. Companies that reorganize around job titles miss the actual work being done. Those that map tasks first, then restructure, see better outcomes.

This granularity helps with compensation too. If 40% of a role's tasks face high AI exposure, does that justify the same salary as a role with 10% exposure? Forward-thinking companies are already adjusting pay scales based on task irreplaceability.

The index arrives as businesses move from asking "if" to asking "how" about AI integration. Task-level data provides the "how" with unprecedented precision.

JobForesight plans quarterly updates to track shifting exposure patterns. For hiring managers navigating AI's impact on their teams, having open, granular data beats guessing every time.