CFO best practices in the evolving generative AI era

OpenAI’s $6.6 billion funding round could be a Pyrrhic victory for the company whose technology has become synonymous with generative artificial intelligence — one that could ultimately serve as a warning for the burgeoning GenAI industry at large.

In early October, the Microsoft-backed business closed a long-anticipated funding round which left it flush with the cash to train up its latest GenAI models and kick the tires on new products, with participation from investors like Nvidia, Thrive Capital and Softbank bringing its valuation up to $157 billion, according to the company. Even as one of the largest funding rounds by a private company closed, however, investors and industry experts were already discussing whether the $6.6 billion would be enough to cement OpenAI’s future long-term.

While the ChatGPT creator reported a spike in its revenue growth over the past year, it also expects to lose approximately $5 billion by the end of the year — with losses related to expenses including the costs of running its products, according to financial documents seen by The New York Times.

The business has also experienced turmoil in its leadership team during the funding round. While CFO Sarah Friar was a top driver for the funding round, top OpenAI executives including its chief technology officer and two senior researchers departed the company in September, days before its funding round closed. Additionally, OpenAI faces growing competition for its ChatGPT products from businesses with deep pockets that rival its main backer Microsoft — which also participated in the most recent round — like Google parent Alphabet and Mark Zuckerberg’s Meta.

The funding round exposed several concerns with OpenAI’s business, with questions arising surrounding funding, potential challenges to revenue and profitability, return on investment challenges and lingering issues regarding data ownership and privacy, experts told CFO Dive. As the best-known, and likely the best-funded, GenAI company, the hurdles OpenAI is facing could mean more than just stormy seas for the company itself: they could indicate dark clouds for the overall GenAI space.

For many companies, the task of divining which way the wind is blowing will fall to their CFOs, tasked with deciding where to spend their companies’ funds — and for many finance chiefs, GenAI’s shine has been dulling for months.

The funding wheel keeps turning  

To be sure, the spotlight on GenAI has yet to dim significantly. Apart from the $6.6 billion OpenAI just pocketed, funding for AI startups reached $19 billion in the third quarter, representing 28% of all venture dollars, according to Crunchbase data. Meanwhile, the number of U.S. companies willing to invest $10 million or more in GenAI is set to double in 2025, CFO Dive recently reported, citing data from Ernst & Young.

But while the funding environment has remained strong, investors — and the CFOs looking to potentially utilize the technology — are starting to ask deeper questions of the companies in their sight lines before drawing out their checkbooks.  Many companies have popped out of the woodwork “positioning what they had been doing, using the more recent nomenclature around AI, machine learning, etc,” said Mark Schwartz, IPO and SPAC advisory leader for Big Four firm EY. “But how much are they doing that's different, relative to …some of the algorithmic-type stuff that they have been doing in the past?”

Understanding that separation is critical for investors as they mull which AI-driven companies are deserving of their capital. “Investors are very keen to understand the technology, to understand if it's truly new, truly novel, and truly potentially revolutionary, as opposed to more of the same,” Schwartz said in an interview.

The need to isolate truly revolutionary companies is beginning to be reflected in venture funding: The $19 billion funneled into AI startups represents a drop from the $23.4 billion seen in the prior quarter, according to Crunchbase, while notably, deal flow for Q3 declined 22% to 947 from the 1,211 rounds seen in Q2.

That “likely means more big deals for more proven AI startups (or at least those that can convince investors of their AI chops) and fewer early seed and Series A deals for young startups,” Crunchbase reported.

“I think underneath the giants there, there's a lot of due diligence that's happening,” Schwartz said of how investors are thinking about GenAI. Some companies are being marked as those that have a differentiator, which are still seeing robust attention in the private markets, but, “there are other companies that, as investors are looking under the hood, they are struggling,” he said.

Crucially for those startups, the world’s largest GenAI player may be proving to be a lackluster poster child. With the company expected to log a $5 billion loss this year, its $6.6 billion round may just be one in a long potential line of raises — sopping up venture dollars and creating more wariness for investors approached by other AI-driven companies.

In fact, given its current burn rate, it’s likely OpenAI will need to complete a similar funding round in 2025 — at a higher valuation than its current $157 billion, tech reporter Ed Zitron predicted in an Oct. 2 newsletter.

“I would imagine, if you're a startup trying to throw AI out there to get huge funding rounds, you’re pissed at Sam Altman,” Matt Cotter, CEO of AP automation software provider Pairsoft said. “Because those guys keep cashing checks, and now everybody turns and looks at you and says, ‘Wait a minute. They're already there, and now I can see their financials.’”

ChatGPT, show me the money

Thanks to its new funding, OpenAI now has access to approximately $10 billion in liquidity, comprised of the $6.6 billion round and a $4 billion credit facility the ChatGPT operator inked with participating banks including Goldman Sachs and JPMorgan, the company announced. OpenAI is also anticipating its revenues will continue to spike, estimating it will hit $11.6 billion next year — compared to expected annual sales of $3.7 billion this year, The New York Times reported.

However, that revenue growth goes hand in hand with rising costs, which represents a key challenge as the company targets that $11.6 billion in annual revenue.

GenAI companies such as OpenAI also face expanding costs related to customer capture, computing power and infrastructure costs. Computing power remains one of the largest cost sinks for OpenAI, and it increasingly needs more than even its top backer Microsoft may be able to supply — Microsoft has not supplied the company with enough power quickly enough, leading OpenAI to approach other potential sources like Oracle, according to reports. 

Part of the issue is that the company’s main cost driver is also its main revenue generator. Presently, the company’s biggest generator of revenue are the subscriptions it sells to its ChatGPT products, with about 10 million users shelling out about $20 per month for such access — a price the company expects to hike to $44 over the next half decade, the NYT said.

“I think what somebody's got to figure out how to do, and I don't think anybody's figured it out yet, is how to change the ratio,” Cotter said. “Revenue keeps doubling and tripling and it's super attractive, but so does cost, because we haven't cracked the code on, how do I deploy all this infrastructure. How do I deploy all this computing power? How do I deploy all this talent?”

Overall, OpenAI expects ChatGPT to generate $2.7 billion in revenue this year. Increasing its subscribers for ChatGPT would mean more revenue, but the sheer computing power needed to generate queries for those users will quickly turn new users into cost centers, Zitron wrote in his Oct. 2 newsletter.

Its other main source of revenue is selling its products to enterprises, where it expects to generate $1 billion by the end of this year. However, that enterprise figure is perhaps the biggest black mark against the company’s future, and by extension, against the monetary draw of the technology itself.

“This suggests a fundamental weakness in the revenue model behind GPT, as well as a fundamental weakness in the generative artificial intelligence market writ large,” Zitron wrote. “If OpenAI cannot make more than a billion dollars of revenue off of this, then it’s fair to assume that there is either a lack of interest from developers or a lack of interest from the consumers those developers are serving.”

Rising prices here could also worsen what Zitron has termed as a coming “subprime AI crisis,"  where “thousands of companies have integrated generative AI at prices that are far from stable, and even further from profitable,” he wrote in a Sept. 16 newsletter.

The attention paid to OpenAI has also prompted companies to funnel more of their budgets into the technology, driving up capital expenditures in a way that may not be sustainable for those smaller companies. 

OpenAI “has been a central, key foundational linchpin to the AI thesis with ChatGPT being the "iPhone Moment" for the world with the introduction of generative AI to enterprises and consumers,” Dan Ives, managing director and senior equity research analyst at Wedbush Securities said in an Oct. 3 note.

“Now we are seeing a $1 trillion of AI Capex over the next three years as OpenAI has been the linchpin to AI success and adoption we are seeing at Nvidia, Microsoft, Google, and across the tech world,” he wrote.

Keeping the eye on the ROI prize

The close attention investors, industry leaders and technologists are paying to OpenAI’s revenue model speaks to a growing need for specifics, rather than potential.

CFOs and their C-suite fellows want both a clear use case from the technology, and, more notably, ways to ensure they’re seeing a positive return on investment from their AI spending. Analysts like David Cahn, a partner at Sequoia Capital, have urged caution in the AI gold rush, with those who remain “level-headed” able to build stable companies, CFO Dive previously reported.

For investors, “what we're hearing is at a super high level, there is a lot of, I'll say, unknowns with respect to just how powerful and just how revolutionary this can be,” Schwartz said.

“I think that this AI revolution has the potential to transform anything and everything top to bottom. I think that right now, we are in the discovery phase of trying to figure out how impactful it can be,” he said. However, “the lack of clarity on valuations in the market” currently is indicative that the space is still in a nascent phase, where leaders are still figuring out how to best move forward, he said. 

For finance chiefs, the GenAI conversation has steadily shifted away from integrating the technology less like a silver bullet, and more like a patch —  something that can solve a particular issue and is being developed accordingly. Indeed, “a focus on practicality, a focus on targeted solutions is going to be, for the world that I live in…a strategy that wins the day,” Cotter said.

Remaining level-headed is one thing — squeezing the ROI investors want to see out of GenAI, in the time frame they want to see it, may be quite another, especially as GenAI companies themselves face challenges growing their revenue where they need to be. Developing those targeted solutions takes a lot of time, although “eventually they're going to be stickier,” Cotter said. “Eventually you're going to get a premium price for them. The problem is, they don't happen in a month.”

Even as leaders isolate specific use cases and projects, “I still don't think that argues that somebody's cracked the code of, ‘how do you change those ratios of revenue to cost?’” Cotter said. “If I'm taking in $6.6 billion in funding, ‘give it time’ is probably not the argument that anybody wants to see here.”

Data coal for the GenAI fire    

OpenAI is already facing a number of challenges, but its last hurdle may be the one that sees it stumble the hardest: data. While the company is facing profitability challenges, the lack of a tidy profit isn’t necessarily a death knell for a business, especially in the technology space; the “move fast and break things” approach proved effective for companies like Uber, which rode to success on wave after wave of venture capital before ever posting a quarter in the black.

But for every Uber, there’s a WeWork: while OpenAI could potentially attract more funding, giving it the time it needs to change its cost to revenue ratio to something more balanced, it could face a supply issue. Simply put, to do all of that, the company needs data. 

ChatGPT and other large language models need access to hordes of data to grow, but gaining access to it isn’t as easy as it might sound. For one thing, with the current training needs of LLMs, “the industry’s need for high-quality text data could outstrip supply within two years, potentially slowing AI’s development,” the Wall Street Journal wrote in an an April report. 

Taking aside the issue of sheer supply, GenAI companies also face an issue accessing much of the data that is available now. Many companies don’t keep their information in formats that are easily accessible for AI tools, meaning they need to set aside more time and money to re-format that information. Furthermore, OpenAI has already been slammed with multiple lawsuits surrounding its access of protected information, raising questions of data privacy and ownership as GenAI usage expands.

“Data provenance is going to be the thing that is existential to OpenAI,” Cotter said. “It is the coal that fuels the plant. if it were to get cut off, legally, technically, something like OpenAI is dead…it thrives on data, period. The process for getting it cut off, though, is murky at best.”

Still, the data question is just one more open-ended problem OpenAI needs to solve to lay the fears of investors to rest. Ultimately, the fate of OpenAI could also be divorced from that of the broader GenAI space, with other players rising to steal the “iPhone moment” away from the ChatGPT creator.

But for finance chiefs, what the cracks here could indicate is that a close, detailed audit of GenAI may be past due — and maybe, one that should be conducted the old-fashioned way.

Microsoft is expanding its collection of tools for automating business functions such as the office of the CFO with the launch of new artificial intelligence “agents” capable of learning and executing certain tasks autonomously, the software giant said Monday.

The company is introducing 10 new autonomous agents, including two designed for finance professionals, in its Microsoft Dynamics 365 enterprise resource planning platform, according to a Monday blog post penned by Bryan Goode, corporate vice president at Microsoft. Pricing and licensing details will be shared once the new agents enter the “public preview” stage later this year, a spokesperson said via email.

The company sees a “huge opportunity” for using such agents to automate corporate finance workflows, according to Georg Glantschnig, Microsoft vice president of Dynamics 365 ERP Applications. “There are still many manual processes” within the finance function, he said in an interview, adding that areas like accounts payable and receivable are particularly ripe for automation.

The move comes as the AI innovation race among enterprise software providers is heating up. It follows Salesforce’s recent introduction of a suite of autonomous AI agents known as “Agentforce.”

Microsoft’s latest rollout builds on a May announcement in which the company described its vision for independent agents that can “automate and orchestrate complex, long-running business processes with more autonomy and less human intervention.” It’s part of a growing trend in which tech vendors are promising the next stage in generative AI applications — beyond tools focused primarily on information retrieval or insights delivery, according to Rowan Curran, a senior analyst at Forrester.

“Buyers are rightly excited about the potential of agentic AI systems but the reality of implementation is going to be just as challenging, if not more so, than the current generation of advanced RAG [retrieval augmented generation] systems,” Curran said in an email. “Having a strong data foundation will be essential for building useful AI agents. Data quality and management aren't problems that can be swept under the rug.”

Next month will mark the two-year anniversary since OpenAI introduced its generative AI-powered ChatGPT tool, which grabbed the world’s attention with its ability to answer all kinds of questions in human-like fashion, draft reports, compose songs and poems, and write software code, among other activities.

Soon after announcing a multi-billion-dollar investment deal with OpenAI in January 2023, Microsoft developed Copilot, its own ChatGPT-like chatbot, which has since been added to a vast array of products and services offered by the software giant.

“Our next big evolution is this agent concept,” Glantschnig said.

As part of its Monday announcement, Microsoft said it’s rolling out a series of new AI agents for sales, service, finance, and supply chain teams. 

A new account reconciliation agent for Microsoft Dynamics 365 Finance automates “the matching and clearing of transactions between subledgers and the general ledger, helping them speed up the financial close process,” according to Goode’s blog post. The company also introduced a new financial reconciliation agent designed to help teams prepare and cleanse data sets.

Sponsored content

By OneStream

Artificial intelligence (AI) has become integral to many industries, revolutionizing how businesses operate and make decisions. In the finance function, AI is no longer a futuristic concept — it’s becoming ubiquitous. AI is now embedded into daily workflows to enhance efficiency, accuracy, and decision-making capabilities. From predictive analytics in forecasting to anomaly detection during financial closes, AI is transforming the way in which finance teams work. This transformation is enabling finance to navigate today’s complex and fast-paced business environment with more confidence and agility.

Today, finance teams are tasked with managing vast amounts of data, maintaining compliance, and providing strategic insights to guide business decisions. Traditional methods of managing these tasks often rely on manual processes that are time-consuming and prone to error. AI is instead stepping in to automate many of these routine activities, allowing finance professionals to focus on more value-adding tasks such as strategic analysis and decision-making. In short, AI is becoming an essential tool for finance teams striving to stay competitive.

Enterprise Finance AI: OneStream's Commitment to the Office of the CFO

To speed the adoption of AI for finance, OneStream is committed to ensuring the Office of the CFO can obtain material value from AI and GenAI development. OneStream's approach to AI, called Enterprise Finance AI, focuses on empowering finance teams to fully leverage AI’s potential. This commitment improves workflows, ensures accuracy, and delivers meaningful insights.

Enterprise Finance AI includes three critical areas AI must meet to speed adoption and significant return on investment for the Office of the CFO:

• AI must be trusted

Finance leaders need to trust the data being used. With Enterprise Finance AI, OneStream ensures both first-party and third-party data are verified and provide 100% transparency into how the insights and answers were formed. This verification and transparency ensure that AI models built on this data are reliable and can be trusted to provide accurate insights.

• AI must be secure

Security is paramount in the finance world, especially when dealing with sensitive financial data. OneStream’s AI solutions are built on a secure, enterprise-grade platform, providing finance teams with assurance that data is safe and compliant with regulatory standards.

• AI must be purpose-built

AI is most powerful when seamlessly unified into existing workflows. With OneStream, the AI solutions are designed specifically for finance workflows. This purpose-built design allows finance teams to take advantage of sophisticated AI models without needing to be data science experts.

The Power of Quantitative AI and Generative AI

OneStream’s Sensible AI portfolio provides a set of purpose-built AI solutions tailored for finance teams. These solutions include the following:

1. Sensible Machine Learning (ML) enables AI financial forecasting and AI-powered scenario modeling. Sensible ML allows finance teams to simulate real-time “what-if” scenarios, using the company’s own enterprise data to evaluate the potential impacts of various market conditions (Figure 1). Such conditions include interest rate changes or supply chain disruptions.

• Sensible GenAI is a purpose-built chatbot inside the OneStream platform (Figure 2). The chatbot brings business context to OneStream’s trusted, secure, and financially intelligent data. With Sensible GenAI, finance no longer needs to complete tedious and time-consuming tasks such as creating reports, researching, and entering commentary to explain variances. Instead, those tasks have been reduced to a simple conversation between a user and a financially intelligent, secure chatbot.

• Sensible AI Library embeds pre-built AI models across all workflows within OneStream. As a result, reporting, planning, and close & consolidations teams can reduce manual and error-prone tasks using capabilities such as AI-powered anomaly detection (Figure 3). This ability enables users to elevate their processes with more advanced, exception-based analysis.

Quantitative AI and Generative AI – A Winning Combination

The combination of quantitative AI and generative AI will create a significant shift in how finance teams operate. With generative AI, finance professionals can interact with their data more intuitively, asking complex questions and receiving insightful answers in real-time. This ability accelerates decision-making processes and helps to reduce the bottlenecks often associated with data interpretation and reporting.

By unifying generative AI into existing finance workflows, OneStream is positioning itself to enhance the performance of every employee in the finance function. This synergy allows for everything from automating mundane tasks to enabling more strategic thinking. In turn, finance teams can operate at their best, contributing more effectively to the overall success of the business.

Between quantitative ML capabilities and generative AI, artificial intelligence will be ubiquitous across functions within the Office of Finance. OneStream will be synonymous with AI. Ultimately, OneStream is showing why artificial intelligence is important in the finance function.

Unlocking the Future of Finance

The future of finance is increasingly AI-driven. OneStream’s Sensible AI Portfolio is not only improving the speed and accuracy of financial processes — but also fundamentally transforming how finance teams operate. By automating routine tasks and providing deep, actionable insights, AI is freeing up finance professionals to focus on what matters most: driving business performance to Take Finance Further.

Learn More

OneStream’s Enterprise Finance AI strategy will empower finance teams to perform at their best. With purpose-built applied AI solutions for business planning and with unified generative AI, finance teams now have the foundation to navigate today’s complex financial landscape with confidence. As a result, AI is not only transforming workflows but also enabling finance to drive better business outcomes across the board.

To learn more on why artificial intelligence is important in the finance function, visit the OneStream AI Hub.

To institutional investors, “speculation” is a four-letter word — unless they happen to make a killing. For financial executives, acting on speculation is often a surefire trigger to job loss.

Yet CFOs across several industries are piling billions of dollars into generative artificial intelligence based on fuzzy forecasts of the likely payoff.

“It’s a noisy estimate,” said Daniel Rock, co-founder of AI consultancy Workhelix, referring to common projections of the gains from generative AI. “It doesn’t mean trying is not worth it.”

Generative AI has helped CFOs finely tune customer service, refine forecasting, speed software updates and upgrade other tasks.

Yet as AI brings “the industrialization of knowledge production,” financial executives face a challenge accurately forecasting the return on investment from new, powerful efforts to squeeze more value from data, according to Laura Veldkamp, a finance professor at Columbia University’s Business School.

The challenge often breaks the usual yardsticks for ROI: Data by its nature is harder to observe and price than assets that undergird the industrial era economy such as buildings and employees, Veldkamp said at a symposium on productivity held by the New York Federal Reserve in February.

“There’s a lot of things in this economy in general that we are going to have a hard time measuring and that we just don’t know enough about yet,” Prasanna Tambe, an associate professor at the University of Pennsylvania’s Wharton School, said at the symposium. “Measurement remains a challenge.”

Despite ambiguity, the stampede into generative AI has swept up investors, financial executives and information technology companies of all sizes, CFOs and AI experts said.

“Speculative frenzies are part of technology, and so they are not something to be afraid of,” David Cahn, a partner at Sequoia Capital, said in a June research note. He sees a $600 billion gap between what companies are spending on AI-related infrastructure and the revenue needed to justify such outlays.

“Those who remain level-headed through this moment have the chance to build extremely important companies,” Cahn said. “But we need to make sure not to believe in the delusion that has now spread from Silicon Valley to the rest of the country, and indeed the world. That delusion says that we’re all going to get rich quick, because AGI [artificial general intelligence] is coming tomorrow.”

Financial executives willing to chance an outlay in generative AI, the CFOs and AI experts said, can follow seven tips to cut through the forecasting haze and seize the potential payoffs:

1. Approach ROI Flexibly

A rush into any “new, new thing” cries out for CFO skepticism. Yet when a potentially terrain-shifting technology emerges, top financial executives would do well to consider more flexibility in gauging ROI.

“As a CFO, the first thing you’ll say to someone coming to you for money is, ‘What’s the ROI?’”  Glenn Hopper, CFO at Eventus Advisory Group, said in an interview. “For something like AI, it’s very difficult to determine.”

Greater efficiency is often one of the most immediate payoffs, he said. AI can trim by as much as three days the time between delivery of a good or service and payment, or the day-to-sales outstanding, he said. It can also streamline just-in-time inventory control and avert errors.

“It’s hard without any prior knowledge to evaluate the potential return on investment,” EY-Parthenon Chief Economist Gregory Daco said in an interview. Identifying with precision the potential payoffs for each company sweeps away some of the fuzziness on potential gains.

“We work on a case-by-case basis with a number of clients to say, ‘OK, in your specific sector, for your company with X market position and an X total addressable market, we believe that AI could have this this type of return on investment if applied across these functions,” Daco said. “It’s really a case-by-case approach because every industry and company is different.”

Many of the companies that succeed with generative AI will embrace risk and experiment despite uncertainty over ROI, Rock said in an interview.

“If you get failure, that doesn’t necessarily mean you made a bad choice,” Rock said. “You can cut it off and re-route it to something else.” Rock founded Workhelix with Erik Brynjolfsson and Andy McAfee.

C-suite executives this year have shown flexibility in their expectations for ROI from generative AI, KPMG found in a June survey.

During the first quarter, 51% of business leaders expected the biggest gains in the next 12 months to come from higher productivity, with 47% expecting higher revenue as the No. 1 benefit, KPMG said.

The rankings flipped in Q2, with 52% of 100 respondents ranking higher revenue as the leading improvement, while greater productivity fell to 40%, according to KPMG.

2. Focus on longer term gains

Getting generative AI up and running often demands sizable up-front investment that could otherwise go to other operations with more certain, immediate returns, Rock said.

“There’s opportunity costs to build up something intangible, like a new way of doing things, a new organizational structure or new workflows, and it seems like you’re putting a lot in to get nothing out,” he said. “Later, there’s a sense that ‘Hey, we’re getting a lot more productive’ — you start to get tangible benefits.”

Paradoxically, CFOs most knowledgeable about AI — its capabilities, limitations and costs — tend to expect the benefits over three to five years, Daco said. Their peers with a weak grasp on the technology “tend to believe the benefits will be more immediate — in two years.”

3. Create a ‘wholesale’ strategic plan

Before considering potential return on investment, a CFO should help create a strategic plan for streamlining operations and increasing growth by weaving generative AI into the organization, Daco said. Bolting the technology onto existing operations often yields meager returns.

“It’s not just about saying, ‘Well, we can automate or augment certain functions,’” he said. “It’s about having a wholesale perspective on how to integrate AI into the fabric of an organization rather than just simple add ons.”

As with any strategic plan, goal setting from the start is essential, according to Adriana Carpenter, CFO at Emburse, a provider of expense management software. “It’s always best to start with the question: ‘What is the outcome that I’m trying to achieve?’” she said.

While creating a broad strategy from a high-elevation viewpoint, CFOs should also help ensure that employees, all the way down to front-line staff, will make the most of generative AI, the CFOs and AI experts said.

“You need to develop a corporate culture that is AI savvy, and think about how to refine your workforce in a way that allows you to put into place a top-down, strategy perspective with a bottom-up, operational perspective,” Daco said.

4. Prioritize your AI projects

A CFO could apply generative AI to most company functions but should resist that temptation and prioritize instead, the financial executives and AI experts said.

“You probably want to stack-rank the different opportunities that are there because there's lots of them and you definitely have limited resources,” Rock said.

Along with alignment to strategy, a generative AI project should draw on existing staff, focus on an operation that they will find compelling and offer the prospect of a quick success, he said.

Before launching a proof of concept, “pick something that you know will be successful and will change minds,” Rock said. “When you’ve demonstrated ROI, then you can say, ‘Cool, now we can move on.’”

Some early adopters of generative AI, although confident in eventually achieving company-wide gains, focused first on picking a surefire project rather than risk a failure that would sour employees on the technology, Rock said.

“You have to prioritize the business outcome you want to achieve,” Carpenter said.

Generative AI has proven especially potent in boosting revenue by analyzing customer data and monetizing it, Carpenter said. It also can help identify where a company is underpricing compared with the competition or how it can expand its customer base.

5. Befriend your CTO, CIO

As the holder of the purse strings, the CFO may have a tense relationship with the chief information officer or chief technology officer, who may not always get funding for their top projects, the AI experts and financial experts said. A CFO should set a friendly, collaborative tone.

“The CFO and CIO — in general —  their areas are historically considered as cost centers, so they’re both fighting with each other,” Hopper said. “It’s important to collaborate together on this to gain efficiencies and really understand that AI is a strategic capability that’s going to strengthen the financial performance and operational excellence of the company —  if they work together on this, everybody wins.”

The CFO also needs a deeper understanding of technology than in previous decades, the AI experts and CFOs said.

Previously, the CFO would have largely focused on how AI and automation would help the controllership, the back office and other tasks within the finance function. Now a CFO needs to cast a broader net and understand the value of generative AI across the entire company.

“I’m not saying you have to become a data scientist or developer, but you need to understand your business, your industry so you have this breadth of knowledge,” Hopper said. “If you don’t know how the information is created, how will you be able to correct it?”

A tech-savvy CFO will also sooner identify high-return innovations in generative AI.

“There’s all sorts of new breakthroughs occurring — every week there’s some new fascinating thing,” Rock said. “You need to stay on top of new developments and build your own detector for the impossibly high-value innovation for your company.”

6. ‘Define’ the data

As part of accurate ROI measurement, the CFO needs to ensure a precise, company-wide definition of the data fed into generative AI, the AI experts and CFOs said. Otherwise, different parts of the company may use the same data to make inconsistent conclusions, undermining the technology’s value.

“In the early days you ended up with these data silos because everybody was doing their own citizen development and building their own data fiefdoms,” Hopper said. “If you’re only taking a slice of the pie at a time, you’re not going to get maximum value. The way you’ll get maximum value is through a shared data warehouse and using it across the company.”

CFOs should keep in mind that the costs of launching a generative AI application — including creating a “data dictionary” and detailed governance of information as part of “responsible AI” — may exceed estimates, Carpenter said.

7. Bring employees along with you

CFOs seeking to achieve a high ROI will err by deploying generative AI primarily to cut payroll, the AI experts and CFOs said. Instead, the technology offers a potential tool for streamlining operations and lifting employee job satisfaction, they said.

“It’s a massive mistake to view this as a cost-saving technology in terms of workforce rather than an augmenter that enables people to get a ton of work done and focus on things that are genuinely hard to do,” Rock said.

Large-language models such as ChatGPT will imperil jobs for workers in dozens of professions, including accountants, auditors, financial quantitative analysts, blockchain engineers, interpreters, mathematicians and journalists, according to a study by researchers at the University of Pennsylvania and OpenAI.

The technology over time will streamline at least 10% of the tasks performed by 80% of workers, and half of the tasks done by 19% of workers, the researchers said.

To dispel worker anxiety, CFOs should “cultivate an AI-savvy culture with a desire to move things forward,” Daco said. They should ensure that employees are “on board with the transformation and that fears of job loss are alleviated.”

Mastering the ability to precisely gauge ROI from generative AI will not happen in a matter of months, the financial executives and AI experts said. To succeed, CFOs will need a balanced mix of resolve and patience.

“There are a lot of learnings that need to occur, there’s going to be fits and starts and it will take time for each type of business in different industries to evolve and accelerate the adoption of this,” Carpenter said. “But generative AI is here to stay.”

Call it the generative AI paradox.

CFOs are chasing the hope of a bonanza from generative artificial intelligence in one of the most frenzied investments in technology in decades.

Yet they have, at best, a loose grip on how to measure the innovation’s ultimate business payoff, according to financial executives and AI experts.

“I haven’t seen anything quite like this,”  said Daniel Rock, who co-founded Workhelix, a generative AI consultancy, with Erik Brynjolfsson and Andy McAfee.

Since the public launch of ChatGPT in November 2022, “there has been this ‘Ah-ha!’ moment, when people say, ‘This is capable of things that I didn’t think information technology systems could do,’” Rock said in an interview, noting the prospect that AI will streamline operations, better serve customers, upgrade forecasting, speed software programming and improve marketing and sales.

The CFO rush to generative AI will likely prove especially rapid because adoption is happening primarily from the bottom-up — from consumers — rather than top-down — from businesses, San Francisco Federal Reserve Bank President Mary Daly said.

The release of ChatGPT marked a “surge moment,” as “generative AI went from a thing others did to relative ubiquity overnight,” Daly said at a University of Chicago conference in March. “This moment led to what I think of as ‘people-driven diffusion.’”

As a result, CFOs and their C-suite colleagues face shareholder pressure and rising expectations for AI-enabled efficiencies and customer service, Daly said. They must move “from apathy to action” faster than with other formative innovations including steam power, electricity and the internet.

“Our conclusion is that you will essentially see the benefits of generative AI over the next five years,” EY-Parthenon Chief Economist Gregory Daco said in an interview. To hit a similar payoff, computers took 10 years, electricity more than 40 years and steam power 80 years, Daco said.

Rapid adoption of generative AI will enable companies to better innovate, harness the new skills of employees, retain customers and gain market share, Daly said.

At the same time, quick deployment of generative AI requires CFOs to measure ROI faster than prior technologies or risk wasteful spending, according to financial executives and AI experts.

AI shakedown

CFOs must also look out for hype and outright fraud, according to the top cop at the Securities and Exchange Commission.

“While perhaps not quite yet a perfect storm, there’s certainly one brewing around AI,” SEC Enforcement Director Gurbir Grewal said. “Every day, we see revolutionary technological advancements in artificial intelligence, or AI, that promise to transform nearly every aspect of our lives, including our financial decision making.”

“Every day, we see individuals, corporations, analysts and others touting these developments,” he said in an April speech. “Every day, we see companies attempting to not only develop AI capabilities, or harness it to improve their productivity and growth, but also to attract and retain investors.”

Most investors (61%) believe that rapid adoption of AI is very or extremely important to a company’s value, according to a PwC survey.

In turn, CFOs and their C-suite colleagues harbor “fear of missing out,” Glenn Hopper, CFO at Eventus Advisory Group, said in an interview. “They’re getting pressure from investors and boards — at every minute, from everybody.”

Consumers have also set a high bar, according to financial executives and AI experts.

“At some companies AI deployments will be something that must be done because customers expect it,” according to Rock, who is also assistant professor of operations, information and decisions at the University of Pennsylvania’s Wharton School. “It’s just going to be table stakes.”

During the coming year, 37% of U.S. companies plan to invest at least $100 million in generative AI, KPMG found in a survey of 220 companies published in March.

Spending on AI software will surge to $298 billion by 2027, 140% more than in 2022, according to Gartner, with 35% of investment focused on generative AI.

“You’re not going to be able to stop the momentum,” Hopper said.

Nvidia, arguably the decade’s top momentum stock, leads the juggernaut behind generative AI adoption. Its share price during the past year has rocketed 153%.

Yet a hint of speculation by Nvidia on the operational value of its products suggests that the onrush into the new technology may, at some point, venture onto shaky ground.

Countries as well as companies have hired the company to upgrade $1 trillion worth of data centers into “AI factories to produce a new commodity — artificial intelligence,” Nvidia CEO Jensen Huang said in a quarterly earnings call on May 22. “The next industrial revolution has begun.”

Every $1 invested by a cloud service provider in Nvidia’s software and networking can, over four years, yield $5 in revenue, Nvidia said in a “Company Overview” on a page titled, “The High ROI of High Compute Performance.”

In a footnote, however, Nvidia said in the May report that its calculation is merely an “illustrative example.”

To be sure, CFOs gauging ROI in generative AI are trying to map fuzzy terrain, the financial executives and AI experts said. Some payoffs from the emerging technology are potentially illusory, especially in the short term.

Only 5.2% of organizations attributed more than 10% of their earnings before interest and taxes to use of generative AI, McKinsey said in May, citing a global survey of 1,363 executives.

Big Data, Big Challenge

The challenge of measuring the value of data predates generative AI, as well as the advent of so-called Big Data and advanced analytics two decades ago.

CFOs and investors have struggled for years to precisely value proprietary data on customers, markets, products and other information.

The measurement challenge did not stop the meteoric rise of Amazon, Google, Uber and other Silicon Valley giants that rely on detailed, self-generated data to fuel profits and growth.

The path-breaking companies rode an explosion in the use of the internet, smartphones and other tools for connectivity and data sharing. Investors place more value on their proprietary data than on their buildings, workers or technologies, Laura Veldkamp, a finance professor at Columbia University’s Business School, said at a New York Fed conference on productivity in February.

CFOs in industries that arose before the advent of so-called information technologies often face a bigger challenge gauging a return from generative AI across their operations, Rock said. “I’m not sure that ROI in gross terms is going to be there.”

AI cost estimates by some companies have fallen short of the mark by as much as 1,000%, Nisha Bhandare, a Gartner vice president and analyst, said at a Gartner CFO conference in May. “Given how new AI is, we don’t really know how much it costs, and are learning as we go.”

Editor’s note: This is the first of two reports on the challenge of measuring return on investment from generative AI. In part two, CFO Dive will describe how financial executives can limit the cost and maximize the payoff from the technology even though, at the outset, they may lack a solid estimate of the potential gains.

Companies that deploy generative artificial intelligence in financial reporting face 12 audit risks that range from flaws in governance and compliance to failures in preventing fraud and cyberattacks, according to the Center for Audit Quality.

“It is critical that companies have a governance model in place that enables them to know where and how these technologies are being used in their organization,” Dennis McGowan, vice president of professional practice at CAQ, said Friday. Given the rapid rise of generative AI, “we’re only beginning to see — and understand — how the use of this technology might be impacting financial reporting,” he said.

Auditors for companies that use generative AI often confront a “black box” challenge when they can neither interpret nor explain how the technology generates information, CAQ said. The problem grows when “financial reporting processes and ICFR [internal control over financial reporting] become more sophisticated and outputs from the technology are unable to be independently replicated.”

One out of three auditors see companies in their industry deploying or planning to deploy AI in financial reporting, CAQ found in a 2023 survey. The proportion will likely grow as companies explore how AI “can streamline or enhance accounting and financial reporting,” the center said.

Accountants are increasingly using generative and other forms of AI to prepare technical accounting memos, to smooth routine tasks such as writing Excel formulas or footing financial statements and to analyze large volumes of data for unusual transactions or unauthorized system access, McGowan said.

Financial executives also use AI to summarize contract terms, prepare the initial suggestion for the accounting treatment of a transaction, and analyze data for budgeting or for understanding variances and costs, he said in an email reply to questions.

“CFOs see the opportunity that generative AI presents, including the streamlining of processes and systems and opportunities for innovation,” McGowan said. “What is less clear is their awareness of risk,” even though audit committees now regularly include AI on their meeting agendas, he said.

CFOs should consider focusing on 12 hazards from generative AI, CAQ said, including:

• Governance – the failure to identify and manage AI applications throughout a company;
   
• Regulation – use of generative AI in ways that violate regulations, laws or contracts;
   
• Skills – employees lack the knowledge to oversee or use generative AI effectively and safely;
   
• Fraud – management, employees or third parties use generative AI to commit or conceal crimes;
   
• Data privacy – confidential data is erroneously entered into a generative AI application;
   
• Security – generative AI is vulnerable to cyberattacks, the intentional insertion of flawed data, or deliberate efforts to prompt bogus conclusions from the applications;
   
• Flawed selection or design – the choice of a generative AI application that does not achieve the desired objective;
   
• Error-prone foundation model – the company adopts an unreliable large language model that generates inaccuracies or biased information;
   
• Flawed training – faulty training of the generative AI model generates repeated output errors;
• Weak performance - due to inadequate testing, the generative AI application “hallucinates,” or provides incomplete, inaccurate, unreliable or irrelevant information;
   
• Defective prompts – employees fail to ask generative AI accurate questions, yielding unintended or irrelevant information;
   
• Inadequate monitoring – after deploying generative AI, companies fail to closely track output to ensure the technology is functioning as intended. 

“Generative AI has the potential to be transformative to business,” McGowan said.

“CFOs have a role to play in assisting their organizations in establishing governance structures that are set up for success,” he said. “They can do this by establishing a strong foundation by linking the governance over AI to their organizations AI strategy and by building literacy among employees around emerging technologies.”

Product enhancements and productivity gains are among the top benefits that business leaders expect to achieve from their planned investments in generative artificial intelligence, according to the results of a recent survey by KPMG.

The vast majority (97%) of executives indicated plans for GenAI investments over the next 12 months, with 43% expecting an allocation of $100 million or more. When asked what they saw as the greatest value creation that might come from these investments, half of the respondents said improving products and services via customer data analysis, while 48% said “enhancing efficiency to generate greater productivity.”

“We’re entering the next phase of GenAI, moving from pilots to transformational programs,” Steve Chase, KPMG’s vice chair of AI & Digital Innovation, said in a press release. “Early experimentation has proven the potential of GenAI, sparking a readiness for greater investments that will deliver enterprise-wide productivity gains, reshape business models and create new revenue streams.”

Despite the growing frenzy around GenAI, many companies are taking a measured approach toward adoption, according to Sanjay Sehgal, head of markets at KPMG. 

“They want to make sure they do this in a thoughtful way,” Sehgal said in an interview.

Given the size of the companies polled by KPMG, the dollars that are being invested into GenAI are “probably in line” with the types of outcomes that executives are hoping to achieve at this stage — if not low in some cases, he said. KPMG surveyed 220 U.S.-based business leaders representing organizations with an annual revenue of $1 billion or more.

The proportion of S&P 500 companies mentioning AI on earnings calls rose to 36% in the fourth quarter of last year, an all-time high, Goldman Sachs analysts wrote in a February research note shared with CFO Dive.

The report highlighted several companies, including Microsoft, Amazon, and IBM, that are expecting AI to produce results such as enhanced productivity, reduced costs, and improved product offerings.

In one example, IBM CFO James Kavanaugh told investors during a January earnings call that the company remains “laser-focused on our productivity initiatives as we digitally transform our business processes and scale AI within IBM.”

“This includes simplifying our application and infrastructure environments, streamlining our supply chain, aligning our teams by workflow, reducing our real estate footprint and enabling a higher value-added workforce through automation and AI-driven efficiencies,” Kavanaugh said.

KPMG predicts that GenAI performance indicators will continue to evolve over time, with early productivity boosts giving way to more transformative investments. However, enabling long-term business value first requires “meaningful investments” in areas such as data security, governance frameworks and workforce preparedness, the firm said in its release.

“When you put all of that together, that’s an investment that companies have to make to then ultimately be able to reap the value that’s going to be on the other side,” Sehgal said.

Top GenAI investment priorities currently include workforce training and capability building as well as developing governance programs to implement the technology in a responsible way, according to the release. Over half (58%) of organizations said they plan to provide mandatory GenAI skills training in the workplace. 

The research comes on the heels of a Deloitte study showing mixed views from CFOs on the outlook for GenAI investments. That poll found that finance leaders are planning to spend modestly on GenAI next year as they navigate some of the risks and uncertainties surrounding the technology.

Thirty percent of the Deloitte survey respondents indicated either uncertainty about the appropriate metrics to use for gauging the value of their investment in GenAI or a lack of current measurements. Those responding to a question about top barriers to GenAI deployment within the finance function in particular cited technical skill gaps, adoption risks, and culture and trust issues, among other concerns. 

Digital transformation was a top focus for companies in 2023, especially following the launch of OpenAI’s ChatGPT generative artificial intelligence tool — which sparked a flurry of investment among businesses looking to capitalize quickly on the new technology’s potential.

However, it’s critical for companies not to fall into the trap of investing in technology simply for the sake of it — rather, companies should be approaching digital transformation and technology integrations with a CFO mindset, said Raphael Ouzan, founder and CEO of engineering and technology talent platform A.Team. That means creating an outcome-focused digital transformation strategy, rather than assessing company-wide implementations that can often fall short of their lofty goals.

“When you think about product engineering… you think outcome first,” Ouzan said in an interview. “And outcome, that's the ROI language of the CFO.”

Taking the CFO approach   

Companies have continued to place digital transformation high on their priority lists, but many digital transformation initiatives are driven by a need to respond to hype rather than a clear business strategy — a pitfall contributing to the failure of over half of these initiatives, according to a September 2023 study by KPMG.  

The GenAI spending spree exemplifies the challenge faced by many companies, which need to keep pace with emerging technologies, but also need to ensure they are driving efficiencies as they come out of an uncertain economic environment.

“On one hand, you have to get to ROI very quickly, you need to increase margin in an inflationary environment,” Ouzan said. “On the other end, there's an acceleration of the pace of technology that we've never experienced in our lifetime.”

Applying a CFO point of view throughout a company’s digital transformation strategy can help achieve that balance, said Ouzan, who founded the New York-based A.Team in 2020. Taking an outcome-focused approach where companies can zero in on individual teams — bringing in skilled experts as needed — can potentially provide greater benefits, “so instead of doing those massive projects, you could do some things that are much smaller and get to much more transformation, much faster,” Ouzan said.

“When we come in, we come in basically with the CFO mindset of, what are the biggest problems right now,” he said. “And we have this approach which is basically, you don't jump to product innovation. You start with efficiency, because that is the first most immediate value proposition and business case for AI.”

Keeping the focus on efficiency

CFOs themselves also play a crucial role in helping their organizations respond to these types of challenges, “because ultimately CFOs know better than most how to value and how to put a premium on optionality,” Ouzan said.

However, many CFOs are working with budgets that are likely to remain flat or potentially decline given economic uncertainty, making it key to keep an even closer eye on the organization’s cost centers and where they will be spending to innovate. The key issue for CFOs tasked with balancing efficiency and innovation in this type of environment will be to learn what Ouzan called “iterative agility;” being able to swiftly experiment with new technologies in a way that creates value.

“It's important to realize that there's a discovery element to this, that you cannot foresee all future revenue and risk of a particular technology project or product capability building until you actually do it and iterate on it,” he said.

The buzz around artificial intelligence has only intensified in recent months, leaving business leaders with the unenviable task of separating the technology’s potential from its reality,

While determining AI’s true applications is a must, the larger question CFOs and accounting leaders need to ask themselves is, “do you truly understand how AI is going to affect the people you hire…as well as those that are going to serve your customers?” said Susie Duong, senior director of research and thought leadership at the Institute of Management Accountants.

A CPA, Duong joined IMA two years ago. Her previous roles include a three-year stint at professional education firm Becker, and she has also served as a visiting assistant professor of accounting at University of Michigan’s Ross School of Business and an assistant professor of accounting for the NUS Business School at the National University of Singapore, according to her LinkedIn profile.

Making AI welcome

While they don’t occupy the top seat when it comes to driving generative AI strategies at their companies — a study by Gartner found CFOs were fourth on that list, behind chief technology officers, chief information officers and CEOs — finance chiefs are nonetheless intimately involved in the decision process when it comes to how and where to incorporate such technologies.

CFOs are also displeased with how digital investments across their organizations are performing, Gartner’s study found. Implementing AI is pricey, Duong pointed out; if businesses are choosing to incorporate it, “at the same time, you have to give up something else that you could have done,” she said. “This is the opportunity cost.”

To make that cost worth it, finance leaders need to ensure they’ve developed a forward-looking AI strategy that provides a return on investment or strategic benefits for the business. CFOs face challenges in four major areas when it comes to bringing AI into the finance and accounting space, according to a recent study by IMA authored by Duong: understanding the technology itself, ethics and governance, operational issues and lastly, the human element.

To take full advantage, finance leaders will need to foster seamless collaboration between the technology and their employees the IMA study found. While AI won’t replace humans, it will “tremendously impact the future of work for professionals in accounting and finance,” the report argues, and CFOs need to be fully prepared for that future. That means educating both themselves and their employees on the technology itself.

“Working with AI is more like working with a new colleague,” Duong said. “To build the trust, to some extent, you have to understand the algorithm behind it.”

That might be challenging for finance and accounting professionals, who are not typically trained on the technical side, she said. It’s important when bringing AI into the business for CFOs to “pay more attention to your people, at all levels,” she said.

“Tell them exactly why you're doing what you're doing, and why this is beneficial to them, not just to the organization,” Duong advised when it comes to educating one’s employees about the company’s use of AI. “And also provide plenty opportunities for upskilling, or re-skilling your employees.”

Bringing one’s workers into the AI conversation is critical not only because it can help smooth the technology’s way into the company, but also because workers remain wary over the potential threat AI represents to their jobs, Duong said.

In the short time since ChatGPT was launched by OpenAI in November 2022, AI job loss has already abounded, with 37% of business leaders claiming the technology replaced workers in 2023, according to a recent CNBC report. Large tech companies like Google, meanwhile, have laid off hundreds of staff members as they continue to invest further in AI, stoking replacement fears.

Leaders should “invite your employees into this whole process, especially the function that is going to be affected,” Duong said. That way, they feel as though their opinion is valued, which can help alleviate some of their fear of replacement as well as make it more likely for them to use the AI technologies that are incorporated, she said.

Prepping for the new accounting age

Finance chiefs also can’t rest on their laurels when it comes to AI education: such a thing has to be done continuously to ensure the company and its talent are keeping pace with the rapid evolution of this and other emerging technologies, Duong said.

It’s also important for CFOs to take a forward-looking approach when it comes to how AI and employees will interact with each other as the technology becomes more commonplace in the finance and accounting sector.  

Finance chiefs are already dealing with a shortage of talent in the accounting world, and while they hunt for skilled talent — or new tech to fill the gap — those that are aspiring to enter the accounting world have different expectations than the old guard, Duong said.

Today’s college students are already learning how to use AI as part of their work, Duong said — and when those students graduate and enter the accounting profession, “their perception of accounting and finance will be completely different,” she said. Finance chiefs need to consider such changing perspectives in an AI-driven world, she said.