How AI is Transforming Public Sector Citizen Services

With artificial intelligence (AI) technologies expected to contribute over $20 trillion to the global economy by 2030, there’s no question that AI is here to stay. This means millions of businesses are now looking for ways to digitise and incorporate these tools into their operations.

While many of the leading private companies now have well-established AI processes and protocols in place, it’s taken a little longer for governments and public sector businesses to get things going.

Nevertheless, we discovered that up to 73% of government interactions are now digital. It’s clear that government bodies and departments are well on their way to taking on the unique risks and challenges posed by AI adoption.

In this article, we’ll take a look at how governments are beginning to leverage AI to their benefit. We’ll also offer some of the specific advantages that AI provides, as well as explore some of the challenges and ethical concerns that it brings.

Much of the data used as supporting evidence comes from our 2025 Connected Government Report. Be sure to check out the full report for key insights into AI and how public sector organisations are using it to drive change.

• Why AI matters for the public sector
• Key applications of AI across public services
• Benefits of AI in the public sector
• Real-world examples
• Challenges and ethical concerns
• The future: agentic AI and citizen engagement
• FAQs

We live in a world where a large majority of people are immersed in the digital landscape. Many are already using AI services and technologies in both their private and professional lives. This is especially true for younger generations who have grown up in this environment.

As such, there is a very high level of expectation from many individuals regarding public sector and government services and how they operate.

AI has the power to vastly improve public sector operations to more effectively meet the needs of its citizens. The use cases are virtually endless , from renewable energy to healthcare delivery to processing vast swathes of paperwork that notoriously clog every government agency. Public safety agencies are deploying AI to support personnel, streamline operations, and analyse large datasets instantly.

Working on the frontline of any government department is tricky at the best of times. You’re constantly expected to handle caller frustrations and complaints and resolve challenging issues.

The rise of AI agents, usually in the form of chatbots or automated telephone services, has dramatically reduced the pressure on public sector workers. They are now capable of resolving most common queries and questions by automatically directing users to the most appropriate resource.

They are also available to offer 24/7 support. Gone are the days when an individual would have to wait until the following morning to get their problem resolved. It can now be done out of hours at the user’s convenience. Additionally, businesses can empower their chatbots and voice assistants with multilingual capabilities, removing language barriers altogether.

And with agentic AI slowly being rolled out to create a holistic, cross-departmental approach to problem-solving, members of the public are likely to see a reduction in wait times and increased efficiency as the technology becomes more refined.

Salesforce, Connected Government Report

Many government and public sector agencies have already turned to Salesforce. With Agentforce, departments are able to streamline and automate several core functions through the use of intuitive AI agents and an agentic AI infrastructure, with complete control and observability over all processes at all times.

While human agents are capable of working at a reliable and steady pace, many administrative tasks, such as data entry and document verification, can be accomplished through automation, driven by emerging technologies.

Governments can also use technologies such as predictive analytics to help anticipate where there might be increased demand for public services, so they can allocate resources appropriately.

A common frustration for many constituents is the time it takes for things to get done, and the burden on government IT workers is well-documented. And they acknowledge the issue, with 83% of workers suggesting that they are expected to do more to meet rising expectations.

So – anything that is able to speed up the process and remove this frustration should be implemented wherever possible.

At Salesforce, our MuleSoft platform is the ideal solution for busy government departments. When used in collaboration with our Agentforce system, it is able to offer dynamic interactions with data to identify bottlenecks in workflows and deliver seamless automation.

A common issue with legacy technology is the build-up of huge data silos packed full of information that is difficult to decipher and organise. This is likely to be data that government departments would’ve needed at some point, and may well need again, which is why using AI to integrate systems and enable cross-department and human-AI collaboration is essential for modern government agencies.

According to our research, large public sector organisations use an average of 973 apps for their day-to-day operations, yet only 28% of them are fully integrated. This is a key indicator that while governments are becoming more digitised, there is still a long way to go before they can offer their citizens fully integrated, efficient services.

Every day, public servants navigate a tangled web of disconnected systems.
Manual processes, outdated technology, and overwhelming backlogs make it nearly impossible to access the right data at the right time.

Citizens themselves are ready for change. Research shows that 91% of respondents in Australia and New Zealand are willing to share their data with the government if it leads to better services.

Public servants want to meet these expectations, but the tools available to them often fall short. Without integrated, accessible data, the gap between citizen needs and government capabilities widens—undermining trust and service excellence.

By breaking down unwieldy data silos, government agencies are able to organise their data effectively to inform real-time decision-making processes and develop a unified 360-degree view for proactive communication.

An AI program is only as good as the data it’s provided with. That’s why it’s vital that, when governments do use AI programs, they are developed with data that has been checked and verified.

This is another area where public sector workers feel that more needs to be done, with only 28% feeling that their data is ready to support AI technology.

At Salesforce, our DataCloud platform is vital for creating a single, standardised database where agents can access key datasets at a moment’s notice. Our customers can take raw data from all of their sources across the internet and turn it into information that is harmonised and actionable.

With access to high-quality, verified data, governments can really start to reap the benefits of AI. There will be far less risk of bias, with AI programs able to successfully detect errors and inconsistencies. Additionally, data-driven decision-making is an excellent way of building trust and transparency between a government agency and the people it serves.

The adoption of AI across public sector agencies and businesses can be tricky to navigate. Many workers are still understandably nervous that AI agents are here to replace them , and customer confidence in AI-driven technologies and utilisation can be a little shaky in terms of trust.

However, agencies that can facilitate the implementation of AI into their operations are likely to see the following benefits:

• Automation: Workers will no longer be required to carry out the laborious manual tasks that can cause backlogs in the system, and they will be supported with things such as case or call summation and data-driven recommendations.
• A smarter service delivery: Virtual assistants will be able to offer round-the-clock support to users, and approvals, verifications and decision-making will all be completed much quicker than before.
• Improved transparency and trust: Public sector businesses will be able to offer much more personalised services in areas such as healthcare. Government processes can be made much more transparent, with clear insights into how an individual’s data is being used.
• Enhanced collaboration: With the assistance of agentic AI, departments will be empowered to collaborate much more effectively with shared, standardised data protection and access, and operational processes.

With all this in mind, it’s clear that a government agency with full AI adoption will be in a much better place than one who is yet to make this transition.

It’s encouraging to see that, while we are still at the early stages of AI adoption, there are dozens of agencies and public sectors across the world that are already thriving in this new digital government environment. Below are a few of what we believe to be the most effective uses of AI within the public sector.

Launched in 2018, Elements of AI is a series of free online courses designed to introduce AI basics to non-specialists in the public. The courses include self-study resources, interactive content and task-based work that teaches concepts, usage methods and limitations of the technology.

The Finnish government pledged to educate at least one per cent of its population, a target that was quickly met and surpassed.

Our government has also been at the forefront of public sector AI adoption for some time now. Services Australia uses AI to develop automated processes, including the use of Optical Character Recognition (OCR) to digitise written forms of evidence and Interactive Voice Response (IVR) systems to guide customers towards the most appropriate resource or help channel.

They are also highly transparent in their use of citizen data, with adherence to Australia’s AI Ethics Principles and Policy for the Responsible Use of AI in Government clearly stated.

Another government agency that has taken positive steps towards adopting AI into their infrastructure is Transport for NSW . They have recently introduced Asset AI, a powerful digital tool designed to assist with their road maintenance operations. Using data from sensors installed on certain vehicles, they are able to gain vital insights into road conditions and highlight potential defects such as potholes and cracks.

The long-term goal is to be in a position where they can take proactive steps in road maintenance long before issues are able to escalate.

With AI still being a relatively new entity, it’s not surprising that there are still some issues and challenges that the public sector is likely to face.

Some of the main challenges include:

• Transparency and explainability: Members of the public have the right to understand how their data is being used, how services function, and how an AI tool came to its decision. Failure to offer adequate explanations related to these points runs the risk of entering a ‘black box’ situation, which will usually exacerbate concerns around fairness and accountability.
• Bias and fairness: A government’s job is to cater for all its citizens, regardless of social and demographic factors. It’s essential that a public agency trains its responsible AI programs using reliable, clean data that won’t distort the program’s functionality.
• Privacy and security: Governments will be in possession of huge amounts of sensitive citizen data. It’s essential that they deploy robust cybersecurity features and protocols to make sure that this data is protected at all times.
• Skills gap: The average citizen is still likely to be unaware of what AI is, what it does, and how it works. This goes for public sector workers, as well. Government agencies must ensure that their workforce is well-trained in AI programs if they are to use them to create effective public services. Programs such as Trailhead from Salesforce provide educational resources related to AI, as well as a whole host of other topics and subjects.

A final challenge that governments need to overcome is the continued lack of public trust in AI and concerns around its ethical use. Governments will need to convince their constituents that their services are both safe and ethical. Transparency is an effective way of achieving this.

Another way is to invest in technologies such as the Einstein Trust Layer from Salesforce. It is designed to provide businesses and agencies with a robust set of guardrails that will protect the privacy and security of citizen data. Features such as data masking and zero data retention are included as standard, with the Layer designed to help bodies remain compliant with IRAP and ISO standards.

AI within the public sector isn’t just about faster processes. It’s about transforming how citizens interact with government services by empowering them through information, enhanced administrative practices, and improved service delivery. All citizens, no matter where they are located, want their governments to make their lives better and to work for them.

Trust is a vital part of AI in government, meaning data protection, integrity, and security is critical. It creates an opening for public sector organisations to establish foundational components (including analysing and standardising large datasets, implementing enhanced security measures, and demonstrating innovation in data modelling) to set them up for an AI-focused future.