Responsible AI can revolutionize tax agencies and improve citizen services.

The new era of generative AI has fueled the exploration of AI use cases to increase productivity, improve customer service, increase efficiency, and scale IT modernization.

A recent study commissioned by IBM® found that a whopping 42% of enterprise-size companies surveyed had actively deployed AI, and an additional 40% were actively exploring the use of AI technology. However, potential risks have slowed the public sector to explore AI use cases and deploy new AI-based tools.

But according to IBM Institute for Business Value’s latest CEO study, 72% of government leaders surveyed said the potential productivity gains from AI and automation are so great that they must take significant risks to remain competitive.

Driving innovation for tax agencies with trust in mind

Tax or revenue management agencies are part of the public sector that can benefit from responsible use of AI tools. Generative AI could revolutionize tax administration and move us toward a more personalized and ethical future. However, tax agencies must adopt AI tools with appropriate oversight and governance to mitigate risks and build public trust.

Although these agencies face many complex challenges that are unique to each country, most share the goal of increasing efficiency and providing the transparency that taxpayers demand.

In the world of government, the risks associated with AI deployment present themselves in a variety of ways, and the stakes are often higher than in the private sector. Mitigating data bias, unethical use of data, lack of transparency or privacy violations is essential.

Governments can help manage and mitigate these risks by relying on IBM’s five fundamental properties for trustworthy AI: explainability, fairness, transparency, robustness, and privacy. Governments can also create and implement AI design and deployment strategies that keep humans at the core of the decision-making process.

Explore the views of global tax leaders

To explore the perspectives of global tax authority leaders, the IBM Center for The Business of Government is collaborating with the American University Kogod School of Business Tax Policy Center to organize a series of roundtables with key stakeholders on AI and modern taxation. Drawing on insights from academics and tax experts around the world, this report helps these organizations understand how they can leverage technology to improve efficiency and deliver a better experience for taxpayers.

The report details the potential benefits of expanded use of AI by tax agencies, including enhancing customer service, speeding up threat detection, identifying and responding to tax fraud, and enabling citizens to access benefits more quickly.

Since the report was released, follow-up roundtables have enabled global tax leaders to explore the next steps in the journey to bring the world’s tax authorities closer to the future. At both meetings, participants emphasized the importance of effective governance and risk management.

Enhancing taxpayer experience with responsible AI services

According to FTA’s 2023 Tax Administration Report, 85% of individual taxpayers and 90% of businesses now file their taxes digitally. And 80% of tax agencies around the world are implementing advanced technologies to collect taxpayer data, and more than 60% are using virtual assistants. According to the FTA survey, this is a 30% increase from 2018.

For tax agencies, virtual assistants can be a powerful way to reduce wait times to answer citizen inquiries. 24/7 assistants, such as watsonx™’s advanced AI chatbot, can help tax agencies by decentralizing tax assistance and reducing errors to prevent incorrect processing of tax returns. These AI assistants also help streamline fast, accurate answers that deliver improved experiences along with measurable cost savings. It also allows compliance with a tax-by-design system that provides taxpayers with early warning of collateral errors that, if left unresolved, could contribute to significant tax losses to the government.

However, these advanced AI and generative AI applications come with risks, and institutions will need to address issues related to data privacy and protection, trustworthiness, tax rights, and the hallucinations of generative models.

Moreover, prejudice against marginalized groups remains dangerous. Current risk mitigation strategies (including a human role in the system and robust training data) are not enough. Every country must independently determine appropriate risk management strategies for AI, taking into account the complexities of tax policy and public trust.

What are your future plans?

Whether using existing large-scale language models or creating their own, global tax leaders should prioritize AI governance frameworks to manage risk, mitigate reputational damage, and support compliance programs. This is possible by training generative AI models using our own quality data and having a transparent process with risk mitigation and safeguards to identify and alert on instances of drift and harmful language.

Tax agencies must ensure that the technology delivers benefits, is transparent, unbiased, and produces appropriate results. As leaders of these organizations continue to expand their use of generative AI, IBM can help global tax leaders deliver personalized and supportive experiences for taxpayers.

With decades of experience working with the world’s largest tax agencies and combined with leading AI technologies, including watsonx™ and watsonx.governance™, IBM can help tax agencies scale and accelerate personalized, responsible deployment of governed AI. .

Learn more about how watsonx can help guide government into the future.