AI in Finance and Accounting: Separating hype from reality

For business leaders, the term “artificial intelligence” (AI) can spark a range of questions, like:

How are my peers adopting this technology?
What actual value has been delivered to finance leaders? What does the cost-benefit analysis of AI look like?
Is my organization ready and equipped to take on something like this?

And it is a delicate balance—striving to guide your organization strategically, invest in innovation, and keep true to your short-, mid-, and long-term goals.

We’re hoping to break down the sense of urgency around AI, reconciling the hype from reality with real market data on how organizations are adopting it, highlighting specific use cases, and the timelines and perceived barriers to adoption.

AI beginnings

The release of ChatGPT in November 2022 was a turning point in AI—making even the harshest skeptics consider its impact on society and the way we do business. Now, AI is the primary highlight of news headlines, earnings releases, and product roadmaps, all in efforts to differentiate and build market share.

While the conceptual opportunity is real, the discerning business leader will understand the gap between where firms are positioned today and what it will take to achieve these groundbreaking results for their organizations.

We surveyed our clients on their AI readiness

To provide these insights, Clearsulting administered a survey to current and past clients. Our goal was to gather a comprehensive dataset that would speak to the state of AI in finance and accounting.

Survey participants are leaders within their respective organization’s Finance and Accounting function and make up a sample of various industries and organizational sizes.

Current state of AI in Finance and Accounting

Organizations are taking this seriously. Of those surveyed, 100 percent are adopting AI or plan to in the future, prioritizing audit, data entry, financial forecasting, and reconciliations to start. But when evaluating adoption rates, it’s important to consider the spectrum that exists within the deployment of AI.

The deployment of AI can vary—it’s no surprise, though, that most organizations opt for use cases that are quick and easy to deploy, like using AI personal assistants and the embedded capabilities within enterprise systems: predictive forecasting, advanced cash application, fraud detection, among others. They tend to leave more complex adoption, like building customized models for financial forecasting, anomaly detection, and automated reconciliation, for last.

As you can see below, only 6% of respondents indicated models with customization were in deployment. Across all deployment types, organizations indicated these applications were focused on audit, data entry, financial forecasting, and reconciliations.

Value and impacts achieved to date

Though organizations are still just beginning their journey to AI adoption, many have been able to prove its tangible, qualitative, and quantitative value. Since most organizations have focused on integrating the use of AI personal assistants, increasing efficiency and effectiveness of tasks was a highly reported outcome. Alternatively, improvement in reporting accuracy and decision making were least reported as those use cases typically require more advanced, predictive capabilities.

Once organizations effectively adopt co-pilots and personal assistants, that’s when they begin to expand capabilities, such as with customized AI Agents—created by training out-of-the-box generative pretrained transformer (GPT) models on their own organization’s data. This data includes text data within policies and procedures to financial and operational data in enterprise data lake houses. The more complex the build, the higher the perceived value. For example, AI Agents can help support end-user adoption of a newly implemented financial system, minimize training, and ongoing enablement.

Deep dive

AI agent use cases

From a value perspective, AI agents are extremely adept at helping organizations address diverse operational and strategic challenges. Leaning on the advanced capabilities in automation, natural language processing, and predictive analytics, AI agents can help:

Remove manual effort: AI agents excel at automating low-value, repetitive tasks, like data entry, troubleshooting, report generation, and compliance checks. That gives your team time to focus on higher-value tasks, improving overall productivity and job satisfaction.
Strengthen data-driven decisioning: Acting as intelligent knowledge management tools, AI agents consolidate information from disparate systems and provide actionable insights. They enable leaders to make informed decisions faster and with greater confidence, especially in areas like financial forecasting and anomaly detection.
Support scalable solutions: By standardizing workflows and automating complex processes, AI agents ensure consistency and efficiency across large-scale operations. For example, in global organizations with decentralized operations, AI agents centralize access to critical resources, reduce operational redundancies, and enable seamless collaboration across teams. They’re also designed to be adaptable, learning from new data and user interactions to meet changing organizational needs without increasing costs or complexity.

Looking ahead

What happens beyond initial adoption is uncertain. Only 30 percent of survey respondents reported continued investment in AI projects over the next 12 months. The majority anticipates continued, albeit slower progress, citing uncertainty in adoption readiness.

Barriers to adoption

To adopt advanced, predictive AI will require a convergence across an organization’s people, processes, and technology. Teams will need to unlock the value of interdisciplinary data to tell the full story of an organization and fuel cutting-edge insights.

In Finance & Accounting, this is particularly challenging as teams continue to be burdened by disparate processes and systems that require significant effort to maintain. Traditional finance modernization efforts are complementary to AI—and initiatives like systems rationalization, process standardization, and data infrastructure optimization help teams further streamline and create efficiencies.

Defining success

To be well positioned for the future, organizations will need to prioritize strategy, governance, and collaboration. There was a strong sentiment among respondents for elevating future investment with enhanced oversight and cross-functional governance, such as with councils, steering committees, project teams, and reporting structures.

Stand out organizations are leaning on a comprehensive set of stakeholders across IT, finance, operations, legal, and risk management. They also indicate a desire for external support as part of their adoption strategy, same as with nearly 80% of respondents.

Once chartered, governance functions are focused on defining tangible goals and projects that align to corporate strategy. They will oversee the execution of AI projects, design decisions, cross-functional alignment, change management, and reporting. And for organizations embarking on the next phase of their journey, that includes reporting on the deployment of enterprise data lake houses and data fabrics, master data cleanup programs, systems rationalization, and process standardization as initial priorities supporting advanced AI deployment.

Let’s recap

As a finance leader, guiding your organization’s modernization is more important than ever. Demands to optimize costs, improve data-driven decisions, streamline processes, and foster innovation will continue to increase.

It’s important to recognize that AI isn’t going away—and to reap its benefits, organizations need to deploy it strategically and build momentum long after its initial value has been proven. Leading organizations see maximum success by pairing strong governance and alignment with corporate goals. In parallel, they’re also continuing to prioritize traditional finance modernization to better enable advanced AI use cases.

Want to see what this might look like for your organization?

Let’s chat