OpenAI and PwC: AI Agents to Automate the CFO Function

The timeline of a deal that reshapes finance

On May 4, 2026, OpenAI and PwC made public their official collaboration to bring AI agents into the CFO function. The announcement is not a simple technological integration. On the contrary, it represents a structural redefinition of how large organizations manage the financial cycle.

PwC brings decades of experience in finance process consulting and a global network of enterprise clients. OpenAI provides its advanced language models and the ability to orchestrate autonomous agents. Therefore, the combination yields something unprecedented: a system capable of executing, verifying, and optimizing financial workflows without continuous human intervention.

The timing is not coincidental. In fact, by 2025, the trend toward the automation of cognitive processes in finance had already solidified. This partnership significantly accelerates that trajectory.

What do AI agents concretely do in the financial cycle

The AI agents described in the OpenAI-PwC collaboration operate on multiple levels of the finance process. First, they automate repetitive tasks: accounting reconciliations, monthly closings, and compliance verification. These operations currently consume a significant portion of the time of finance teams in medium and large enterprises.

Furthermore, the agents are designed to improve forecasting. They analyze historical data, macroeconomic variables, and real-time market signals. Consequently, they produce more accurate and up-to-date predictions compared to traditional models based on Excel sheets or static BI software.

A third level concerns internal controls. Agents continuously monitor transactions, identify anomalies, and report potential risks before they become problems. Therefore, the finance team's role shifts towards strategic supervision and interpretation of data, rather than its manual production.

According to research from McKinsey Global Institute, the finance and accounting functions are among those with the highest potential for automation through generative AI. The OpenAI-PwC partnership fits exactly into this space.

Winners, losers, and those who have yet to decide

Who benefits from this agreement? First, large companies that already work with PwC and can access AI solutions integrated into their audit and consulting processes. Among other things, PwC gains a significant competitive advantage over the other Big Four in the race for AI applied to finance.

OpenAI, for its part, is consolidating its presence in the B2B enterprise market. However, the real gamble is demonstrating that its agents function in high-stakes contexts like finance, where errors have concrete and measurable consequences.

Who risks losing ground? Traditional ERP and FP&A (Financial Planning & Analysis) software houses that have not yet integrated agentic capabilities into their products. Similarly, junior finance professionals who perform purely executive tasks might see reduced demand for those specific skills.

SMEs are the ones who still need to decide. These companies don't have immediate access to PwC's enterprise solutions. Likewise, they don't always have the necessary data infrastructure to run AI agents effectively. Therefore, the risk is falling behind as the market transforms.

SHM Studio Reading: The Signal Beyond the News

We of SHM Studio We see this partnership as a market signal rather than an isolated news item. Finance has historically been one of the last business areas to adopt radical innovations. Therefore, when two players like OpenAI and PwC move in this direction, it means that technological maturity has reached a sufficient level for real-world applications.

For Italian B2B SMEs, the question isn't whether to adopt AI in finance processes. The question is when and with what approach. In fact, waiting for enterprise solutions to become accessible without internal preparation means losing precious time.

Specifically, accessible AI tools are already available today that SMEs can integrate into their financial workflows: from automatic expense categorization to narrative report generation, and predictive cash flow analysis. These tools do not require enterprise infrastructure. However, they do require a clear strategy and adequate digital skills.

According to Gartner, by 2027, more than 50% of finance departments in medium-sized organizations will be using at least one AI agent for operational tasks. This figure suggests that the window of opportunity for those who act now is still open, but it is closing.

The construction site is still open: governance and trust in data

The OpenAI-PwC partnership leaves some critical questions open. First and foremost, the governance of AI agents in finance. Who is responsible when an agent makes a mistake in an accounting reconciliation or budget forecast? The answer is not yet universally defined.

Furthermore, data quality remains a non-negotiable prerequisite. AI agents only produce reliable output if fed with structured, complete, and up-to-date data. Many Italian SMEs still have significant gaps in the quality of their financial data. Consequently, investing in data quality is a mandatory step before any agentive automation.

Finally, the theme of regulatory compliance is central. In Europe, the AI Act imposes specific requirements for high-risk AI systems. Financial processes often fall into this category. Therefore, any implementation must be designed with careful consideration of regulatory implications.

Next moves: what's advisable for Italian SMEs to do today

The OpenAI-PwC news doesn't require an immediate response in terms of technological adoption. Instead, it suggests some strategic moves that SMEs can initiate as early as Q2 2026.

• Internal financial data audit Verify the quality, structure, and completeness of the available data. It is a prerequisite for any future automation.
• Identification of repetitive high-volume processes: Map finance activities that consume the most time without adding strategic value. These are the ideal candidates for initial automation.
• Experimenting with accessible AI tools: There are SaaS solutions that integrate AI into finance processes without requiring complex implementations. A limited pilot allows for the acquisition of internal expertise.
• Finance team formation: The change isn't just technological. The team must develop new skills for supervising and interpreting the data produced by the agents.

For SMEs looking to delve deeper into how AI can be integrated into their business processes, the team at SHM Studio — AI Services offers personalized analysis sessions and roadmaps. Additionally, those considering how to communicate their digital transformation can explore our solutions for digital marketing e SEO.

To delve deeper into the topic of AI agents in enterprise contexts, we recommend reading the Harvard Business Review analysis on AI agents in business. It offers a strategic perspective useful for those making decisions right now.

Finally, for those managing their company's B2B communications, our solutions for LinkedIn campaign e Google Ads campaigns can support positioning on innovation-related topics. As well as our service of SEO copywriting can help build editorial authority on these topics. For further details, the team is available via the page contacts or exploring the SHM Studio Blog for further analysis and updates. Furthermore, the section web services offers solutions for those who want to build digital infrastructures adequate to support technological innovation.