The proposed shift in one of India’s toughest exams, the IIT JEE, toward better assessing critical thinking and reasoning is a quiet but important signal of how the future of work is changing; largely under the influence of AI.

That this shift is coming from the IIT ecosystem itself reflects a broader realization: the way talent was historically selected was optimized for a different era than the one we are entering now.

For decades, JEE rewarded extreme early specialization –

A pattern mastery through predictable, syllabus-bound questions.

This worked well in a world that valued stability, clear rules, and execution excellence.

It helped build the technical foundation of a newly independent nation.

But like all strong systems, it optimized for what it measured.

The result was not a lack of intelligence or effort; far from it, but a system that selected for narrow excellence in “kind” environments, where:

• Rules are fixed
• Feedback is fast
• Problems repeat

If you’ve read David Epstein’s Range, he explains why this breaks down in the real world, which is a far more “wicked” environment:

• Rules change
• Feedback is delayed
• Problems are ambiguous

A more dynamic exam pattern flips the question from “Have you seen this before?” to “Can you reason this out now?”
That is the same shift happening in the job market.

AI and agentic systems are already very strong at narrow specialization.

If a task has clear rules, data, and success metrics, AI can increasingly plan, execute, and optimize it.

What does not get automated easily is judgment:

• Problem framing
• Cross-domain trade-offs
• Context
• Human dynamics
• Uncertainty

That is generalist territory.

The difference was never intelligence, it was alignment with what the system rewarded.

As exams become more dynamic and work becomes more agent-driven, the advantage shifts:

• From memorization to reasoning
• From execution to orchestration
• From narrow mastery to adaptable intelligence

JEE changing its pattern is not just about fairness or stress reduction.

It reflects a future that rewards not those who know the most answers, but those who can figure things out when the answers are not obvious.