Why public and private sector AI initiatives will fall short in 2026 without a focus on productivity 

As organizations accelerate AI adoption, a subtle but critical shift is underway. 

Throughout 2025, we saw organizations from across the public and private sectors work to leverage the benefits of AI and incorporate the technology into their operational workflows. 

For government organizations, the focus is largely on using the technology to boost productivity. According to a new report from the OECD, 57% of AI adoption cases in the government support automating, streamlining or tailoring services. 

However, challenges faced in scaling successful AI mean many initiatives remain in the pilot phase. A key contributing factor is a lack of impact measurement frameworks to demonstrate return on investment and thereby prioritize further investment in AI.

Dr. Ranjit Tinaikar, CEO of Ness Digital Engineering, recently highlighted a key insight following extensive conversations with CIOs and CTOs from around the world. 

He advises that AI only delivers value when engineering systems, teams, and decisions are explicitly designed for productivity, not experimentation. 

In 2026, engineering productivity is expected to emerge as a key metric, and will move us away from the experimentation that has dominated the past few years in the industry. 

How AI has rewritten the rules for digital success 

The rapid rise of AI in the past few years was made possible in large part thanks to the pre-existing digital environments that have slowly taken shape. 

Over the past 30 years, organizations slowly switched away from paper-based systems and began relying on digital for communications, record-keeping and everything in between. This digital infrastructure was bolstered even further thanks to the advent of Software-as-a-Service apps that offered targeted solutions for everyday business needs, through to the rise of cloud computing. 

The transition to digital also opened up the doors to a number of new opportunities, such as improved data insights, real-time analytics and more efficient business processes. 

During this period, growth in digital environments was synonymous with scale. In other words, the benefits of the digital system or tool in question would compound as more users were brought on board. 

Organizational leaders measured the success of the project in terms of how widely digital tools were being deployed and how frequently they were being used to boost productivity and increase outputs. Common metrics typically were things like ‘adoption rates’ and ‘time spent in app’.

In turn, this means that these digital environments grew by employing more engineers to build and install an increasing number of applications. 

However, AI is rewriting this equation for good. Now, the real differentiator is not how much an organization builds, but how productively it builds. According to Dr. Tinaikar, in 2026 “AI-native engineering becomes the new norm — Every system will be designed from the ground up expecting AI participation.”

Engineering productivity as a lever for social impact

As AI changes the status quo, organizations worldwide have a chance to drastically increase their efficiency by focusing on engineering productivity. 

The importance of this shift isn’t only important for enterprises. The opportunities for governments and nonprofits are just as profound. Here, productivity is largely tied to human outcomes. 

Further, non-profit and public sector organizations most often don’t generate revenue. This means that resources are limited and largely tied to fundraising or government support. In turn, the way that these funds are used is often closely scrutinized, with huge pressure on civil servants and charity leaders to prove they are delivering impact. 

This means that improving productivity isn’t about efficiency for efficiency’s sake. It’s about helping more people, faster, and with higher quality outcomes. 

Without a focus on engineering productivity first and foremost, AI tools will become another layer of complexity in the digital ecosystem rather than a force multiplier for human outputs. 

Here, we can look to insights from Ness to understand the matter more closely. A study looked at a Fortune 500 manufacturer of engineering equipment that power critical energy and industrial operations worldwide. However, the task of managing the sensor data from thousands of turbines across geographies was becoming increasingly complex.

In turn, data quality, consistency, and AI readiness were unclear, creating hesitation about where and how to invest.

The solution was on practical impact across operations, maintenance, and customer experience that looked to build confidence through feasibility assessments before scaling to make sure that AI can actually move the needle. 

In the same way, this methodology can use engineering productivity as a lever to increase social impact. 

How software development lifecycles impact public and private sector outputs 

The challenges with adopting AI effectively compound for large organizations with complex workflows. Here, the risk that AI will add to complexity rather than boost efficiency is even greater. 

As AI tools enter software development lifecycles, they expose long-standing productivity gaps in legacy software development lifecycles (SDLCs). Here, AI won’t fix broken systems but amplify problem areas such as manual approvals, brittle architectures, siloed data and slow feedback loops. 

For governments and nonprofits, the stakes are especially high. A slow SDLC can have serious real-world consequences, delaying access to benefits, healthcare services or disaster response aid. In turn, poorly built digital public services can erode trust between citizens and the government.

In 2026, inefficient engineering doesn’t just cost money, but also time, equity and impact. 

As a result, traditional output metrics are losing relevance across sectors. Lines of code, feature counts, and project milestones tell an incomplete and often misleading story in an AI-enabled world. 

In their place, more meaningful indicators are emerging. For nonprofits and governments, these metrics translate directly into outcomes such as how quickly aid reaches communities, how resilient systems are during peak times and how confidently leaders can act on real-time data.

When paired with productive engineering models, AI offers a chance to reverse this dynamic by accelerating service delivery while improving reliability and transparency.

Another Ness study that looked at how over 100 software engineers used GenAI for coding found that is had a remarkable impact on the outputs of each user. When engineers worked on routine code updates and maintenance tasks, productivity gains hit 70%. 

Senior engineers saw a 48% improvement across all tasks and a 10% increase in productivity in complex coding scenarios.

For public sector organizations, these gains promise to have a transformative impact. 

Building high-performance organizational functions with technology 

To understand how to solve the issue of poor productivity, we can look to high-performance organizations from the tech industry. Here, according to Dr. Tinaikar, we can see they are reframing the core question they ask of technology teams from “How much did we build?” to “How effectively did AI amplify our people?” 

The distinction is crucial. AI-driven productivity isn’t about replacing engineering with automation but instead about moving friction, improving signal-to-noise in decision-making, and allowing teams to focus on higher-order problems, whether that’s driving business growth or delivering essential public services.

At Ness, that shift is changing how organizations think about the next wave of growth and impact and how digital services are built and delivered through tools like ATONIS, an AI-powered, automated engineering workbench designed to revolutionize the entire SDLC. 

As AI-native engineering models emerge, productivity becomes measurable, repeatable, and scalable by design. 

This approach is equally valuable for the global enterprise that aims to launch cutting-edge innovation to its customers, through to the public-sector institution trying to modernize an aging digital infrastructure. 

By embedding AI across the engineering lifecycle and modernizing operating models in parallel, organizations can ensure technology investments translate into real-world results.

Providing value in the AI era 

What many organizations get wrong about AI transformation is assuming tools alone will deliver value. In reality, success requires rethinking how engineering work is structured, how outcomes are measured, and how decisions are made. 

To move forward in today’s AI-first world, leaders from across the public and private sectors need to align on the new metrics for success, where productivity is the biggest driver of growth. 

In an era where AI is ubiquitous, sustainable advantage will belong to organizations that can consistently turn intelligence into action faster, better, and with fewer wasted cycles. This ensures that value is delivered, whether the goal is keeping shareholders happy or serving society at large.

Article’s featured photo of Dr. Ranjit Tinaikar