Governments Eye AI at Scale but Face Major Execution Barriers: KPMG

A new KPMG report reveals that 48% of governments plan to deploy AI at scale within the next year, but 43% are struggling with execution challenges. Key hurdles include data silos, cybersecurity maturity gaps, and a lack of skilled talent. The report notes that 40% of technology budgets still go to maintaining legacy systems, slowing transformation. Leading governments are shifting focus to citizen outcomes and partnerships to overcome these barriers.

Key Points: KPMG: 48% of Governments Plan AI Scale, Face Hurdles

  • 48% of governments plan to deploy AI at scale in 12 months
  • 43% face execution hurdles in scaling AI
  • Data silos, cybersecurity gaps, and talent shortages are top barriers
  • 86% believe managing AI agents will be crucial in 5 years
3 min read

Nearly half of governments to deploy AI at scale, but face execution hurdles: KPMG report

KPMG report finds 48% of governments plan to deploy AI at scale within 12 months, but 43% face execution hurdles like data silos, skills gaps, and cybersecurity issues.

"Governments report being in the early stages of AI adoption... however, challenges scaling are increasingly evident in the data. - KPMG Global Tech Report 2026"

New Delhi, April 27

Governments across the world are preparing to deploy artificial intelligence at scale over the next year, but many are struggling to move beyond pilot projects due to structural and operational challenges, according to a latest report by KPMG.

The report titled "KPMG Global Tech Report 2026: Government and Public Sector" said that 48 per cent of government organisations plan to deploy AI use cases into production at scale within the next 12 months, even as 43 per cent admitted they are facing roadblocks in scaling their initiatives.

"Governments report being in the early stages of AI adoption... however, challenges scaling are increasingly evident in the data," the report noted.

The findings are based on responses from 115 government leaders and technology decision-makers globally and highlight a widening gap between ambition and execution in public sector digital transformation.

The report pointed out that one of the biggest hurdles is the lack of coordination across departments, with governments "struggling to integrate their data across functions, departments and ecosystems" and being "limited by siloed thinking."

It added that while governments are keen to adopt advanced technologies, much of their spending continues to go towards maintaining existing systems. Around 40 per cent of technology budgets are being spent on maintenance, indicating a slower shift towards transformation.

On cybersecurity, the report flagged a significant gap between awareness and preparedness.

While data security remains a top priority, only 38 per cent of respondents said their cybersecurity capabilities are mature, even as 64 per cent plan to increase investment in this area over the next year.

"Data protection and security concerns are the top factor slowing collaboration," the report said, adding that these concerns are also affecting partnerships on emerging technologies.

The report also highlighted issues around data management, noting that just 37 per cent of organisations have achieved high maturity in enterprise data systems, despite recognising that strong data foundations are critical for digital transformation.

"Clean, reliable, integrated and accessible data is the key to delivering service transformation from technology investments," the report said.

Another major challenge identified is the lack of skilled talent. Governments are finding it difficult to build the capabilities needed to support new technologies, especially AI. The report said 86 per cent of respondents believe managing AI agents will become an important skill within the next five years.

At the same time, governments are increasingly looking to collaborate with external partners to accelerate transformation. About 75 per cent of respondents said they plan to expand their technology ecosystem and partnerships, although concerns around security, costs and alignment remain.

The report emphasised that leading governments are shifting their approach by focusing on citizens and outcomes rather than just processes.

It noted that the most successful organisations are "moving past siloed processes... to instead focus on outcomes, experiences and experimentation."

Overall, the report suggests that while governments are committed to digital transformation, significant gaps in data integration, skills, cybersecurity and execution continue to slow progress.

- ANI

Share this article:

Reader Comments

S
Sarah B
Interesting report. The gap between ambition and execution is universal, not just unique to India. Glad to see cybersecurity being flagged—only 38% mature is scary. We are digitizing everything from Aadhaar to healthcare records, but if the basics aren't secure, we're building on sand. 🏗️🔒
K
Kavya N
That 86% believing AI agent management will be a key skill in 5 years is a wake-up call. Our education system needs to pivot fast—not just coding, but how to work *with* AI. Also, 40% budgets on maintenance? That's like a family spending most of its income on house repairs instead of building a new wing. 🤔
R
Raghav A
Honestly, reading this, I'm a bit worried. Data integration, cybersecurity, skills—all basic foundations but the report says governments are struggling. In India, we have the advantage of a young workforce, but we also have legacy systems in many states. Hope the new AI mission takes these execution hurdles seriously rather than just making announcements.
J
James A
The "focus on citizens and outcomes" shift is spot-on. Too often, government IT projects are about processes and compliance, not user experience. If we can make AI-driven services as seamless as booking a cab or ordering food, that's real transformation. But the skills gap remains the elephant in the room. 🐘
T
Tanvi S
Data security concerns slowing collaboration—this hits home. We have so many govt bodies each with their own data sets (land records

We welcome thoughtful discussions from our readers. Please keep comments respectful and on-topic.

Leave a Comment

Minimum 50 characters 0/50