AI in the public sector: How technology is transforming the relationship between government and businesses

Artificial Intelligence (AI) is no longer a trend. It has become a reality in the Brazilian public administration. According to the ICT Government 2025 survey, released by the Brazilian Internet Steering Committee (CGI.br) in July 2026, AI is already present in 59% of federal and state public agencies and in 27% of municipalities across the country.

For companies operating in tax automation, SAP S/4HANA, and tax compliance solutions, this advancement is more than a statistic. It is a clear signal that the public sector is accelerating its digital transformation, creating new opportunities and requirements for integration between private systems and government platforms.

Judiciary leads adoption and federal Executive branch reaches 70%

The survey reveals that the Judiciary leads the use of the technology, with 94% of agencies already deploying AI in their operations. This is followed by the Public Prosecutor's Office (92%), the Legislative branch (83%), and the Executive branch (55%).

At the federal level, 70% of agencies already use some form of AI application, while in the states this figure is 58%. This movement reflects a structural shift in how the Brazilian government operates and how it connects with the private sector.

Workflow automation is the most common AI application

Among agencies already using AI, workflow process automation is the most common application, adopted by 39% of institutions, with 45% at the federal level and 38% at the state level. This is followed by text mining and language analysis, present in 35% of agencies, and machine learning tools for prediction and data analysis, used by 31% of public institutions.

This scenario has a direct impact on companies that depend on integration with government systems, such as invoice issuance, digital bookkeeping (SPED), tax calculation, and fiscal compliance. As governments automate more processes, real-time interoperability between corporate ERPs (such as SAP S/4HANA) and public platforms becomes increasingly critical.

Generative AI arrives in the federal government

The 2025 edition of ICT Government brought, for the first time, data on the use of generative AI in the public sector. According to the study, 65% of federal agencies are already developing generative AI initiatives in internal processes, while 29% use the technology in services provided directly to citizens.

Additionally, 59% of federal agencies offer courses or training for civil servants, 46% have contracted specific tools, and 43% have established guidelines or manuals to guide their teams.

Municipalities enter the radar and 27% already use AI

For the first time, the survey assessed the use of AI in Brazilian municipalities. The result: 27% of cities used the technology in the 12 months prior to the survey. Adoption varies by population size. Cities with more than 500,000 inhabitants report 85% usage, while state capitals reach 81%. In municipalities with up to 10,000 inhabitants, the percentage drops to 22%.

In municipal administrations, the most frequent applications also focus on process automation (14%) and text mining (12%).

Training is the main bottleneck

Despite the progress, the survey identifies that lack of training remains the primary barrier to expanding AI use in the public sector. Among federal agencies, 19% cite lack of training as an obstacle. In the states, this figure rises to 21%, and in municipalities, it reaches 40%.

Among municipalities, other challenges include AI not being considered a government priority (36%), absence of need or interest (35%), difficulties related to data quality and availability (35%), and high costs (34%).

What this means for tax automation and SAP S/4HANA

For companies operating in tax automation, tax compliance, and ERP-government integration, the advancement of AI in the public sector brings three strategic implications.

The first is the increased demand for real-time integration. As governments automate fiscal, customs, and accounting processes, the need for bidirectional real-time communication between corporate systems and government platforms becomes critical. Solutions based on SAP S/4HANA must be prepared for increasingly sophisticated APIs, web services, and data exchange protocols.

The second implies an opportunity for AI applied to tax compliance. If the government is already using AI for text mining, data analysis, and workflow automation, companies can and should do the same. AI applied to tax calculation, fiscal document validation, SPED inconsistency detection, and tax liability forecasting are applications that are already mature and can be implemented alongside SAP S/4HANA.

The third is the need for data governance and digitized tax documentation. The survey shows that data quality and availability is one of the main obstacles to AI adoption in municipalities (35%). In the private sector, the challenge is similar. Without structured tax data, digitized documentation, and metadata governance, tax automation cannot reach its full potential.

Vottax and the future of AI-powered tax automation

At Vottax, we closely monitor this digital transformation movement in the public sector. Our mission is to connect tax technology, data intelligence, and ERP-government integration so that companies not only meet tax obligations but transform fiscal compliance into a competitive advantage.

If your company seeks to automate tax processes with AI, integrate SAP S/4HANA with government systems, reduce audit risks with intelligent document validation, or digitize and govern tax data to feed AI models, our team is ready to help.

Sources: ICT Government 2025 Survey (CGI.br / Cetic.br / NIC.br), released on July 23, 2026.