Evidence register

Every number we publish, sourced and dated.

We sell AI governance. It would be strange to run a marketing site whose statistics nobody can check. Every figure on this site resolves to an entry below — with the publication date, the fieldwork date where they differ, and whether we read the source ourselves.

2
sources: 12
read at source: 3
as cited: 1
withheld, unverified

Read at source

We opened the document and took the figure from it.

As cited

We read the claim quoted inside a document we did read. The underlying study is named but unopened.

Withheld

Claims nobody here has verified are tracked internally and never rendered on a page.

Source 01

How can the public sector meet the AI moment?

McKinsey & Company, Public Sector Practice

Hrishika Vuppala, Tim Fountaine, Tim Ward, Tony D'Emidio

Jul 2026

Read from PDF, 13pp, retrieved 2026-07-25.

28/100

AI Quotient of the social and public sector — the lowest of 14 sectors measured, against a global average of 33

Technology leads at 44; healthcare scores 34. AI Quotient measures AI maturity across core capabilities and management practices.

Exhibit 1 (McKinsey Tech and AI Quotient Database, 2026)

70% vs 30%

of domain-based AI programs reach production, versus programs led by individual use cases

The strongest available argument for reimagining a whole end-to-end workflow rather than assembling a portfolio of isolated use cases.

p.5

31%

as cited

of public sector employees trust their employer to develop AI safely — against 71% across all industries

Only one in five expects AI to meaningfully affect their daily work.

p.5, citing McKinsey, 'Superagency in the workplace', 28 Jan 2025

as cited

countries now have national data and AI strategies, up from fewer than 20 in 2020

p.5, citing World Privacy Forum and OECD

$1 : $5

as cited

for every dollar spent on AI technology, five must go to change management to capture the value

p.9, citing McKinsey, 'Scaling gen AI in the life sciences industry', 10 Jan 2025

Quoted

“Agencies pulling ahead design for risk management from the start rather than retrofit governance after deployment.”
p.10

“AI run costs can rise quickly in agencies that don't track them, and many agencies don't. … Building financial operations capabilities around AI early is cheaper than retrofitting them after the bill arrives.”
p.9

“A benefit denial, a license revocation, or a public safety dispatch is a consequential decision requiring a human in the loop. A grammar check on a draft response is not.”
p.10

Source 02

The state of AI: How organizations are rewiring to capture value

QuantumBlack, AI by McKinsey

published Mar 2025; fieldwork Jul 2024

evidence 2.0 yrs old

Superseded by

The state of AI in 2025: Agents, innovation, and transformation (QuantumBlack, 5 Nov 2025)

Online survey fielded 16–31 July 2024; 1,491 participants across 101 nations; 42% at organizations above $500M annual revenue. Correlation findings use a Johnson's Relative Weights analysis over self-reported impact — associative, not causal.

Read from web article print-to-PDF, 37pp, retrieved 2026-07-25 (endnotes partly obscured).

71%

of organizations regularly use gen AI in at least one business function

Up from 65% in early 2024. Counting analytical AI as well, 78% use AI somewhere.

Exhibit 9

11%

use it in risk, legal, and compliance — the function that governs AI

Against 42% in marketing and sales. In healthcare, pharma, and medical products the gap is starker still: 63% use gen AI somewhere, 5% use it in risk, legal, and compliance.

Exhibit 10

of executives describe their gen AI rollouts as mature

p.6, complementary survey in a set of developed markets

80%+

of organizations report no tangible impact on enterprise-level EBIT from their use of gen AI

p.17

1 of 25

attributes tested, the redesign of workflows shows the biggest effect on gen AI's bottom-line impact

Associative, not causal — the ranking comes from a Johnson's Relative Weights analysis over self-reported impact.

p.2

28%

of organizations have a CEO responsible for overseeing AI governance

CEO oversight is the attribute most correlated with self-reported bottom-line impact; at larger companies it is the element with the most impact on EBIT attributable to gen AI. A further 17% report board-level oversight, and on average two leaders share the responsibility.

p.1

27%

of organizations review all gen AI output before it is used

A similar share check 20% or less of what their models produce.

p.3

<1 in 5

track well-defined KPIs for their gen AI solutions — the practice with the most impact on the bottom line

Fewer than a third of organizations follow most of the twelve adoption and scaling practices surveyed.

p.6, Exhibit 4

57%

fully centralize risk and compliance for AI — the most centralized element of AI deployment

Data governance follows at 46%; adoption of AI solutions is the least centralized at 23%.

Exhibit 1

13%

have hired AI compliance specialists; 6% have hired AI ethics specialists

p.8

Quoted

“Companies that report capturing value from gen AI are rewiring processes to effectively embed gen AI solutions while appropriately incorporating human-in-the-loop mechanisms to validate models and outputs.”
p.8

Colophon

Quotations are short, attributed, and linked to the publisher. We do not reproduce any publisher’s exhibits, charts, or figures, and nothing here implies endorsement of Ready for AI by the organizations cited.

Spotted an error or a stale figure? Tell us — we will correct it or retire it. Figures are withheld rather than published unverified, so this register is deliberately shorter than it could be.