KPMG AI Quarterly Pulse Survey 2025

Executive Summary

KPMG's eight-quarter longitudinal research tracking enterprise AI adoption, investment patterns, and deployment challenges. Reveals clear shift from experimentation to enterprise discipline across industries.

KPMG's quarterly pulse survey tracks executive sentiment and organizational readiness for artificial intelligence deployment across enterprise segments, providing longitudinal trend data that captures evolving attitudes, investment patterns, and implementation experiences. The 2025 survey reveals a sentiment trajectory characterized by tempered optimism replacing the unbounded enthusiasm of previous quarters, as organizations accumulate practical deployment experience that reveals both genuine productivity improvements and previously underestimated implementation challenges. Notable findings include a widening gap between AI strategy articulation and operational execution capability, growing board-level concern about governance and liability exposure, and increasing recognition that organizational change management represents a more significant adoption barrier than technology capability limitations.

Key Findings

72%

C-suite confidence in AI return on investment rebounded after two quarters of declining sentiment

Of C-suite executives expressed confidence in achieving measurable ROI from AI initiatives within eighteen months, up from fifty-nine percent in the prior quarter's survey.

61%

Talent acquisition and retention overtook data quality as the most cited impediment to scaling AI programmes

Of respondents identified specialised talent scarcity as their top scaling barrier, surpassing data quality concerns for the first time since the survey's inception.

2.4x

Organisations with dedicated AI centres of excellence reported faster time-to-value for generative AI deployments

Faster median time from proof-of-concept approval to production deployment for firms operating centralised AI CoE structures compared to decentralised, business-unit-led approaches.

48%

Regulatory preparedness varied dramatically by geography, with European firms leading and Asian enterprises accelerating

Of Asian enterprises initiated formal AI regulatory readiness programmes in Q1 2025, representing a twenty-two percentage point increase from the previous year.

Abstract

About This Research

Publisher: KPMG Year: 2025 Type: Case Study

Source: KPMG AI Quarterly Pulse Survey 2025

Relevance

Industries: Cross-Industry Pillars: AI Readiness & Strategy Regions: Southeast Asia

Strategy-Execution Gap Dynamics

The survey documents a persistent and widening gap between organizations' articulated AI strategies and their demonstrated execution capabilities. Ninety-one percent of surveyed executives report having formal AI strategies, but only twenty-three percent rate their organizations' execution capability as adequate for achieving strategic objectives. Contributing factors include insufficient data infrastructure readiness, fragmented governance accountability, competing organizational priorities that divert resources from AI initiatives, and unrealistic timeline expectations established during strategy formulation without adequate consideration of implementation complexity.

Board Governance and Liability Concerns

Board-level engagement with AI governance has intensified significantly, driven by regulatory developments, high-profile AI incident publicity, and shareholder governance expectations. The survey reveals that fifty-seven percent of boards now include AI risk as a standing agenda item compared to eighteen percent two years earlier. However, board AI literacy remains insufficient for effective oversight, with directors expressing uncertainty about appropriate risk appetite calibration, liability allocation between organizations and AI vendors, and the governance implications of autonomous decision-making systems operating without direct human supervision.

Organizational Change Management Primacy

Respondents increasingly identify organizational change management rather than technology capability as the primary determinant of AI deployment success. Cultural resistance to algorithmic decision-making, workforce anxiety about displacement, middle management reluctance to relinquish process control, and insufficient investment in user training and workflow redesign collectively constrain AI value realization more effectively than technical limitations. This recognition is prompting budget reallocation from technology procurement toward change management programmes, training initiatives, and organizational design interventions.

Key Statistics

72%
of C-suite executives confident in AI ROI within 18 months
KPMG AI Quarterly Pulse Survey 2025

61%
cited talent scarcity as the top barrier to AI scaling
KPMG AI Quarterly Pulse Survey 2025

2.4x
faster deployment with dedicated AI centres of excellence
KPMG AI Quarterly Pulse Survey 2025

48%
of Asian firms launched AI regulatory readiness programmes
KPMG AI Quarterly Pulse Survey 2025

Common Questions

The gap reflects strategy formulation processes that insufficiently account for implementation prerequisites including data infrastructure readiness, governance framework establishment, talent acquisition timelines, and organizational change management complexity. Strategies developed during peak enthusiasm phases established ambitious objectives without proportionate attention to execution foundations, creating commitments that exceed organizational delivery capability. Closing this gap requires honest capability assessment, phased implementation planning, and sustained executive commitment through the extended timelines that production AI deployment demands.

Board AI governance engagement has intensified substantially, with fifty-seven percent of boards now treating AI risk as a standing agenda item compared to eighteen percent two years earlier. This escalation reflects regulatory pressure, high-profile AI incidents increasing reputational awareness, and shareholder governance expectations. However, board effectiveness remains constrained by insufficient AI literacy among directors, uncertainty about appropriate risk appetite calibration for algorithmic systems, and difficulty evaluating the adequacy of management-proposed governance frameworks without technical domain expertise.