Deloitte's 2026 Global Human Capital Trends survey asked leaders a direct question: is building your organisation's ability to adapt at speed critical to your future? 85% said yes. Asked whether their organisation was actually leading on it, 7% said yes.

That is a 78-point gap between intention and reality — published the same year BCG found that 50 to 55% of all roles will be meaningfully reshaped by AI within the next three years.

The gap has a name. This edition is about what it looks like in practice.

Two MIT Technology Review analyses landed on the same day last month. Both are credible. Both are based on real data. They point in opposite directions.

The first: across the aggregate US labour market, AI-exposed jobs have lower unemployment rates than jobs less touched by AI. Large-scale displacement has not materialised. Only one in five companies is currently using AI in any business function.

The second: workers aged 22 to 25 in AI-exposed occupations have seen a 16% relative decline in employment since generative AI became widespread — even after controlling for other factors. Underemployment among recent graduates has reached 42.5%, the highest since the pandemic. Firms are not announcing redundancies. They are eliminating the junior tasks through which people gain their first foothold in a career, and quietly not replacing them.

Both are accurate. They are not contradictory. They are looking at different levels of the same data.

This is what the talent intelligence gap produces in practice.

The aggregate is calm. The cohort is not calm. Most succession planning, most workforce analytics, and most HR board packs are built on aggregates — average readiness scores, overall bench strength, total headcount trends. The data that would reveal the cohort-level picture exists in almost every large organisation. It is rarely assembled in a form that makes it visible.

BCG's 2026 analysis sets the context: 50 to 55% of all roles will be meaningfully transformed within three years. That transformation is not uniform. The roles changing fastest — often those that have historically served as the entry point to senior leadership pipelines — are precisely where the cohort risk is concentrating, silently, beneath an aggregate that looks fine.

The Deloitte 85%/7% split is not a capability problem or an ambition problem. Leaders understand what is required. They cannot see whether they are achieving it. That is an intelligence architecture failure. And it is the same failure that makes the MIT cohort data invisible to most CHROs — until the bench is already thin.

When your team presents succession health to the board, is the picture shown at the aggregate — or segmented by role family, AI exposure, or career cohort? Would the picture look different at the cohort level?

What evidence would you give your board that your organisation is in the 7% Deloitte found are actually building adaptive capability — not just intending to?

Creating an Adaptable Workforce (Deloitte Global Human Capital Trends 2026) — The source for the 85%/7% finding. The report frames it as an organisational design question; this edition reframes it as an intelligence architecture question. Worth reading in full for the broader context on the adaptive capability imperative.

→ https://www.deloitte.com/us/en/insights/topics/talent/human-capital-trends/2026/creating-an-adaptable-workforce.html

A reality check on the AI jobs hysteria (MIT Technology Review, May 2026) — The aggregate case: AI-exposed jobs have lower unemployment than less-exposed ones; only one in five companies use AI in any business function. The headline that makes the cohort finding below so important to read alongside it.

→ https://www.technologyreview.com/2026/05/26/1137855/a-reality-check-on-the-ai-jobs-hysteria/

It's time to address the looming crisis in entry-level work (MIT Technology Review, May 2026) — The cohort case: 16% relative employment decline for workers aged 22–25 in AI-exposed roles; underemployment among recent graduates at 42.5%. Read alongside the MIT reality check — the two together are the argument.

→ https://www.technologyreview.com/2026/05/26/1137865/its-time-to-address-the-looming-crisis-in-entry-level-work/

AI Will Reshape More Jobs Than It Replaces (BCG, 2026) — 50–55% of roles meaningfully transformed within three years. The succession implication: the role your pipeline is built for today may not be the role that exists in 2028.

→ https://www.bcg.com/publications/2026/ai-will-reshape-more-jobs-than-it-replaces

Edition 8 — The entry-level hiring your organisation stopped in 2026 is the succession pipeline you won't have in 2031 — Edition 8 showed how workforce restructuring creates pipeline gaps. Edition 9 names the intelligence architecture failure that makes those gaps invisible until it is too late. (TTI edition)

→ https://newsletter.totalia.ai/p/the-entry-level-hiring-your-organisation-stopped-in-2026-is-the-succession-pipeline-you-won-t-have-i

The organisations in the 7% can see the cohort-level picture. The rest are managing the aggregate. That distinction is worth sharing.

They can subscribe here: https://newsletter.totalia.ai