As federal agencies in the U.S. face budget cuts, including funding for infrastructure projects and initiatives already underway, they are increasingly expected to do more with less. These are challenging circumstances for government agencies, but behind every new challenge lies opportunity. In this instance, the challenge to do more with less could serve as a catalyst for artificial intelligence transformation, augmenting teams with new capabilities that allow them to boost their productivity and enhance their decision-making.
AI-driven transformation has been well underway in the private sector for some time. Companies have successfully leveraged automation to streamline operations, reduce costs and accelerate project timelines. A recent study by McKinsey took some typical generative AI use cases and modeled them across the private sector economy to assess the technology’s impact — for instance, the application of gen AI to create content such as personalized emails or the ability to carry out context-rich data searches. It found 63 generative AI use cases spanning 16 business functions that could deliver a total value in the range of $2.6 trillion to $4.4 trillion. In McKinsey’s words, “That would add 15 to 40 percent to the $11 trillion to $17.7 trillion of economic value that we now estimate non-generative artificial intelligence and analytics could unlock.”
Now, the launch of ChatGPT Gov has opened the debate about AI in the digital transformation of the public sector. Unlike consumer AI models, ChatGPT Gov is specifically designed for federal, state and local agencies, providing secure, compliant access to AI capabilities within Microsoft Azure’s commercial and government cloud platforms. This could help with adherence to stringent security frameworks such as FedRAMP High, addressing some long-standing concerns around data privacy and governance.
Government agencies, however, face different challenges than private sector companies — from rigid regulatory requirements to complex procurement cycles and the need for accountability and transparency. To unlock AI’s full potential, the public sector must go beyond generic AI solutions and embrace tailored, strategic AI adoption models that directly align with its operational realities.
A smarter framework for AI implementation
A customized, iterative deployment model ensures that AI aligns with an agency’s specific needs rather than forcing operations to conform to a generic one-size-fits-all platform. By embedding AI experts directly within an agency’s environment, organizations can assess real-world inefficiencies, identify automation opportunities, and build solutions that are tailored to their workflows. This approach focuses on understanding the unique operational structure of each agency and applying AI where it can deliver the greatest impact, whether in document processing, compliance reporting, predictive maintenance or resource allocation.
The process should ideally begin with an on-the-ground assessment, where AI specialists work alongside agency teams to identify areas of high value for automation. Once challenges and inefficiencies are mapped out, these experts can then collaborate with leadership to develop AI-driven solutions that fit seamlessly within existing processes. Instead of relying on preconfigured enterprise AI models, agencies can leverage custom-built AI agents — particularly agentic AI, which refers to models that can autonomously make decisions, take actions and pursue goals with minimal human intervention — that are designed with security, privacy and regulatory mandates in mind. Crucially, this implementation should happen incrementally, allowing agencies to integrate AI in phases rather than all at once, ensuring a smooth transition that minimizes disruption.
This measured approach gives agencies the flexibility to refine AI deployments over time rather than committing to a costly system upfront that may not fully address their needs. By gradually introducing AI-driven automation and adapting it based on real-world performance, agencies can ensure compliance, security and long-term sustainability in their digital transformation efforts.
Implementing agentic AI: From T&M to ARR
The adoption of agentic AI in government agencies typically follows a structured pricing transition that balances flexibility with long-term value. The process generally begins with a one-day workshop, which is often offered at no cost, allowing agencies to explore AI’s potential and assess where automation could have the greatest impact. This initial engagement provides decision-makers with a clear understanding of AI’s capabilities before committing resources.
Following this, agencies move into a time and materials (T&M) phase, where forward-deployed engineers are embedded within agency operations to observe workflows, identify inefficiencies and determine high-value automation opportunities. This hands-on approach ensures that AI solutions are developed based on real-world agency needs rather than relying on assumptions or generic automation frameworks.
Once AI-driven solutions are designed and tested, agencies can then transition into an annual recurring revenue (ARR) model, subscribing to custom AI-enabled platforms and agentic AI solutions that continuously evolve based on real-world performance and changing operational requirements. This shift from custom development to a subscription-based service allows agencies to scale AI adoption at their own pace, ensuring that they maintain flexibility while benefiting from continuous improvements in AI capabilities. Therefore, rather than investing in expensive, one-time AI deployments, agencies can tap into AI as an ongoing service, reducing risk and optimizing long-term costs.
Now is the time to embrace AI
Public agencies, more so than private companies, are under pressure to do more with less. Augmenting their existing workforce with agentic AI to boost worker productivity and allow them to carry out complex searches and calculations will be a milestone improvement for the sector. This change is already underway. AI bills have been introduced in more than 40 states, with Connecticut and Texas establishing working groups to assess the use and deployment of AI systems. In the past year, the governors of California, Oklahoma and Virginia have issued executive directives designed to address the operational, IT, workforce and risk dimensions of generative AI.
Make no mistake — agencies that delay AI adoption today risk inefficiency and higher costs tomorrow. However, AI adoption must also be approached with caution and consideration. While off-the-shelf tools such as ChatGPT Gov lay the groundwork for AI-driven collaboration, document processing and policy analysis, custom AI solutions will ensure long-term security, efficacy and adaptability.
Balaji Sreenivasan is the founder and chief executive officer of Aurigo Software Technologies.
Copyright
© 2025 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.
