Document processing
7 Practical Applications of AI in Government
12 min read
—
Jan 19, 2023
Discover the most prominent applications of AI in government and learn how V7 can help government agencies manage large-scale public data across domains.
Haziqa Sajid
In this technology-intensive world, individuals, businesses, and governments use Artificial Intelligence (AI) to automate their workflows and minimize redundant tasks.
In June 2022, Bloomberg reported that AI expenditures of various governments like the US, UK, China, and Canada are increasing. For instance, the US government, in 2020, committed to over $1 billion for AI funding. Similarly, in March 2021, the Canadian government pledged over half a billion dollars to advance its AI initiatives.
The public sector deals with large amounts of data, so increasing efficiency is key., AI and automation can help increase processing speed, minimize costs, and provide services to the public faster.
This article discusses major use cases of AI for governments, including:
Traffic flow analysis
Improving healthcare services
Enhanced large-scale infrastructure monitoring
Processing large amounts of data
Task automation via chatbots
Cyber attack prevention
Smart policymaking
We will also discuss some challenges and setbacks critical to deploying AI in government.
But first, let's take a look at the current state of artificial intelligence in government agencies.
AI in government: Overview
Government agencies have started investing in AI technologies to solve various public sector problems. A Deloitte report on the use cases of AI in government states that
70% of business leaders support AI-driven government projects
80% of government agencies are at the initial or developing stages of reaching digital maturity
By 2024, 75% of governments will have at least three enterprise-wide hyper-automation initiatives launched or underway
A major reason for using AI in government processes is that it can free up millions of labor hours. This can allow government workers to focus on more important tasks and result in the government being able to provide services to the public faster.
A Deloitte survey estimates that automating government workers’ tasks can save around 96.7 million to 1.2 billion labor hours, potentially resulting in annual savings of about $3.3 billion to $41.1 billion.
The deployment of AI in government is not without challenges. There’s a dire need to spread awareness and develop AI expertise among government workers.
Government agencies must also be concerned with the accountability of AI as these initiatives directly impact public and national safety. Deloitte identified key dimensions that the public sector should focus on to build trustworthy AI:
Fair & impartial: Reducing discriminatory, historical, racial, and gender bias from AI decision-making
Transparent & explainable: Helping users understand the AI algorithms
Responsible & accountable: Introducing policies that ensure accountability of AI systems, especially for critical applications
Safe & secure: Reducing cybersecurity risks by adopting DevSecOps approaches to ensure the security of AI systems
Privacy: Ensuring compliance with data and privacy regulations such as GDPR and CCPA
Robust & reliable: Maintaining accurate and consistent outcomes as the AI system scales to a larger audience
These challenges make it more difficult to fulfill budget requirements for AI research and development. However, as AI matures and public accessibility increases, this trend will change over the next few years.
Let's discuss some major AI applications that governments can leverage to improve public sector services.
Traffic flow analysis
World Health Organization (WHO) estimates that 1.3 million people die in road crashes yearly. By effectively applying AI in transportation, governments can significantly reduce road safety issues.
Thanks to technological advancements like computer vision, object detection, drone tracking, and camera-based traffic systems, government organizations can analyze crash data and highlight areas with a high likelihood of accidents. They can employ additional road safety measures to prevent future mishaps.
Moreover, AI and machine learning algorithms provide data-driven analyses that aid officials in controlling traffic flow, preventing accidents and congestion, monitoring logistics, and improving road safety.
Similarly, the U.S. Department of Energy has developed an AI tool called Transportation State Estimation Capability (TranSEC). It uses machine learning to analyze traffic flow, even from incomplete or sparse traffic data, to deliver real-time street-level estimations of vehicle movements.
GRIDSMART is another AI-based system that uses computer vision and object detection to track all moving objects within view. The system uses a fisheye camera – a camera with an ultra-wide angle lens, to track every segment of the road and deliver the information back to traffic managers.
V7’s image annotation and video annotaion tools help government organizations manage high-quality transportation datasets. As a result, agencies can train robust traffic models with advanced monitoring capabilities.
Healthcare and disease spread prevention
AI in healthcare has achieved several breakthroughs in medical science, from early disease detection and prevention to clinical decision support.
AI allows real-time tracking of patients' health. This includes monitoring weight, height, blood glucose, stress levels, heart rate, etc., and feeding this information to AI healthcare systems, which can notify doctors of any possible risks.
Governments can leverage AI to provide effective health services to citizens. For instance, during the pandemic, AI impacted the detection and control of the COVID-19 virus.
In the UK, National Health Service (NHS) formed an initiative to collect data related to COVID patients to develop a better understanding of the virus. Through various partnerships, the NHS set up the National COVID-19 Chest Imaging Database (NCCID), an open-source database of chest X-rays of COVID patients across the UK. This initiative aimed to develop deep learning techniques meant to provide better care for hospitalized COVID-19 patients.
Similarly, the NHS has developed an AI tool that can detect heart disease in only 20 seconds while the patient is still in an MRI scanner. Normally, it would take a doctor 13 minutes or more to analyze the MRI scans of a patient manually.
Likewise, the U.S. Centers for Disease Control and Prevention uses an AI tool to streamline polio virus tracking and reporting. It can identify virus types and form clusters of different disease reports.
The Australian government's syndromic surveillance system, PHREDSS, monitors patient symptoms in hospitals every day to determine emerging disease outbreaks and configure health policies accordingly.
The precision of the AI models is highly dependent on the quality and quantity of the medical images dataset. V7's intelligent labeling tool speeds up the annotation process and provides an end-to-end tool for medical data management.
Large-scale infrastructure monitoring
Governments around the world face difficulty in monitoring real estate. Manual administration is challenging and often proves insufficient in identifying land developments.
These challenges pave the way for the inclusion of AI in government to automate the monitoring and administration of properties.
A French consulting firm Capgemini partnered with Google to develop AI software that can analyze aerial imagery to spot undeclared properties. The software was able to spot 20,000 undeclared pools across France. This discovery helped the French tax authorities collect an additional €10m in tax revenue. Moreover, the authorities say they will use the software to identify undeclared patios, gazebos, and home extensions.
Similarly, in the United States, government organizations and insurance companies use an AI tool to identify any changes in infrastructure or property. An Australian company NearMap has developed an AI tool that provides land identification and segmentation from aerial images.
Nearmap's AI tool is trained on image data that covers 380,000 square miles across the US and Australia.
In summary, AI in government enables authorities to enforce policies that result in better infrastructure monitoring to fight tax evasion and unlawful property changes.
Processing large amounts of data
All public sector departments deal with a lot of paperwork. Manual data entry or verification consumes a lot of time and resources, making it difficult to prove quick services to the public.
A Governing magazine report found that 53% of local government officials cannot complete their work on time due to low operational efficiencies like manual paperwork, data collection, and reporting. As a result, their task backlogs keep piling up, causing further delays in government workflows.
AI-based cognitive automation, such as rule-based systems, speech recognition, machine translation, and computer vision, can potentially automate government tasks at unprecedented speed, scale, and volume.
OCR data entry tools can process large document dumps in minutes, which would otherwise take hours to complete with legacy systems. For example, Georgia Government Transparency and Campaign Finance Commission successfully digitized 40000 campaign finance disclosure forms via OCR.
Task automation: Chatbots
Government officials interact with the public every day to resolve their queries. By replacing these officials, AI chatbots can effectively automate interactions, allowing workers to focus on more complex tasks.
They provide a comprehensive knowledge base for citizens with multilingual support and collect citizen feedback on a large scale.
For instance, the US Army recruitment website uses a virtual assistant, SGT STAR, that has so far answered over 10 million public queries. It guides visitors around the website, answers basic questions, and redirects to a human correspondent when needed.
Similarly, the Department of Homeland Security, USA, uses EMMA, a virtual assistant catering to immigration services. EMMA guides around one million applicants per month regarding the various services offered by the department and directs them to relevant pages and resources.
Likewise, the government of Dubai uses an AI assistant RAMMAS that guides citizens regarding bill payment, application tracking, and job applications.
Cyber attack prevention
Government agencies sit on top of critical public and defense data. This makes them a target for cyber-attacks. Any intrusion in government databases affects national security and damages the public's trust.
AI in government workflows allows agencies to prevent or minimize cyber attacks. Using AI, government organizations can:
Monitor unusual network activity and entry points
Identify potential data vulnerabilities and fortify access restrictions to critical data
Improve the accuracy of intrusion detection systems
Simulate cyber attacks in controlled environments to identify system vulnerabilities
Automate system patching and security updates
As cyberattacks become more and more sophisticated, legacy systems fail to prevent malicious activities.
For instance, Booz Allen identified that common cyber defense tools do not detect intrusion until 200 days after.
The company developed a Cyber AI solution that helps governments and enterprises manage advanced cyber workflows, including
Behavioral modeling
Domain-generated algorithm detection
Malware detection
Credential misuse detection
Microsoft has developed a tool named Cyber Signals which actively tracks 140+ threat groups and 40+ nation-state actors across 20 countries. It helps in intelligence sharing across different government agencies by identifying and tagging malicious entities.
Smarter policymaking
Agencies and policymakers can leverage artificial intelligence to conduct citizen-centric smart policymaking. AI tools provide advanced analytics on public data, allowing policymakers to identify emerging issues related to their regions and constituents.
Researchers Arnaboldi and Azzone have highlighted two significant benefits of analytical information in policymaking:
Increased efficiency and effectiveness in policy planning as it is easier to understand expectations and target major problems
Faster recognition of changing societal trends that allows to modify policies accordingly
For instance, an AI crowdsourcing tool developed by a Belgian technology company CitizenLab was used by Belgian authorities to understand public demands during climate change protests in 2019. As a result, Belgium was able to define 15 climate action priority policies that were curated via public opinion.
AI in government: Risks and challenges
Public-use technologies demand a higher level of accountability and compliance with regulations than technologies developed by the private sector.
This causes some major challenges and risks in adopting AI in government. Let’s go through a few of them.
Lack of AI expertise: High-quality AI resources are rare. A SAS survey states that 63% of respondents believe that they don’t have enough high-quality AI and machine learning resources. Government workers often lack the data science and AI skills necessary to handle the massive scale of citizen data and build state-of-the-art AI models that benefit the public.
Lack of motivation: Typically, government workers are accustomed to their existing workflows. Their existing organizational culture does not encourage an R&D-based environment which is critical for AI adoption.
Challenge of fairness and equality: AI bias has long been a challenge for experts. AI models can display bias based on different characteristics, raising questions about the fairness and integrity of the government. For instance, a facial recognition security and surveillance system used by a government agency could mischaracterize a citizen, leading to false legal action against them.
Concerns of privacy: Collecting large amounts of citizen data may be met with protests from citizens concerned with safeguarding their identities. For instance, contact tracing and self-reporting applications raised some serious healthcare privacy protection concerns during the pandemic. In addition, most government agencies outsource the development of AI systems, which leads to sharing citizen data with third-party companies. Also, it is challenging to get the consent of each individual.
Ethics and lack of accountability: AI models have always been subject to ethical issues, and their public sector implementations are no different. Ethical issues in AI mainly arise due to data biases and privacy. Data biases impact how the model treats certain sects and communities, and the use of public data puts its privacy at risk.
Government agencies must adopt and enforce ethical AI guidelines in different phases of the AI lifecycle to ensure transparency, contestability, and accountability. However, most public-sector AI initiatives are underfunded and understaffed to execute ethical AI policies effectively.
If AI is effectively applied across government departments, it can have a positive impact on the healthcare, energy, economic, and agriculture sectors. However, without clear governmental guidelines, AI can jeopardize the well-being of citizens.
AI in government: Final thoughts
AI holds immense potential for improving government services. With advanced technologies, government agencies can cut labor costs, speed up processes, save man-hours and provide smooth and quicker services to the public.
However, as with any other project, AI adoption poses challenges that the public sector must overcome. Federal offices must start building an AI culture and raise employee awareness. Governments can start with pilot projects, at the same time, pass legislations that facilitate sustainable AI adoption in the long run.
Interested in building enterprise AI applications that facilitate public sector operations? Contact us for a hands-on V7 demo.
And if you want to learn more about AI applications across other industries, check out these resources:
