Artificial intelligence is driving a bull market in stocks, a win for investors. For economists, however, the payoff will be measured longer-term, depending on how billions of dollars of technology investments filter down in terms of productivity and business efficiency.
If they pan out, AI could prove to be more consequential than previous technology waves, and Big Techs are banking on it. Microsoft has invested some $13 billion in OpenAI, establishing a close partnership with the creator of ChatGPT. Amazon has put $4 billion into another generative AI (GenAI) leader, Anthropic. Google parent Alphabet and Facebook parent Meta have signaled their intentions with tens of billions each in planned capital expenditures.
Goldman Sachs Economics Research last year projected total investments as high as $100 billion in the U.S. and $200 billion globally by 2025. These are seen as prerequisites for reshaping business processes and producing major productivity gains, though “the near-term GDP impact is likely to be fairly modest given that AI-related investment currently accounts for a very low share of U.S. and global GDP,” said economists Joseph Briggs and Devesh Kodnani.
In another 2023 study, McKinsey & Co. concluded, “Generative AI’s impact on productivity could add trillions of dollars in value to the global economy . . . the equivalent of $2.6 trillion to $4.4 trillion annually” across 63 use cases that the firm analyzed. (Total U.K. GDP in 2021 was $3.1 trillion, McKinsey noted.) “This would increase the impact of all artificial intelligence by 15% to 40%, [which] would roughly double if we include the impact of embedding generative AI into software that is currently used for other tasks beyond those use cases.
“About 75% of the value that generative AI use cases could deliver falls across four areas: Customer operations, marketing and sales, software engineering, and R&D.”
McKinsey’s value-impact analysis showed banking “to have one of the largest opportunities: an annual potential of $200 billion to $340 billion (equivalent to 9% to 15% of operating profits), largely from increased productivity.”
AI’s potential global economic impact, from McKinsey & Co., The Economic Potential of Generative AI.
It's not just a Big Tech arms race. Accenture announced a $3 billion commitment last year to expand its Data & AI practice and “accelerate clients’ reinvention.” PwC said in April 2023 that a $1 billion AI and client-enabling initiative featured a “relationship with Microsoft, creating scalable offerings using OpenAI’s GPT-4/ChatGPT and Microsoft’s Azure OpenAI Service.”
Such deals are channeling advanced AI into business applications, as is, even more literally, a Bank of New York Mellon Corp. data-and-analytics alliance with Microsoft. It was subsequently announced that BNY was the first major bank to deploy an Nvidia SuperPOD supercomputer along with the high-performance chip maker’s enterprise software “to support the build and deployment of AI applications and manage AI infrastructure.”
Also working to deliver at the enterprise level, targeting $1 billion over three years for banking, government and other key verticals, is analytics-technology leader SAS.
“Financial services is an especially important industry for SAS. Financial services firms are typically at the forefront of technology adoption,” noted Stu Bradley, senior vice president of risk, fraud and compliance solutions at Cary, North Carolina-based SAS.
The financial industry is “often at the forefront from a risk perspective, in that the fraud threats that impact banks are frequently the same or similar to ones that later impact insurers, retailers, telcos and others,” Bradley added. “The investments we make in helping banks overcome their challenges help organizations in other sectors downstream.”
Stu Bradley of SAS
Of specific interest to risk professionals is expanded large language model (LLM) and GenAI capabilities of SAS Viya. The enterprise decisioning architecture “is at the heart of how we help our customers extract more value from their AI investments,” Bradley said.
PwC partner Robin Stein pointed to an agreement announced in May making PwC OpenAI’s first ChatGPT Enterprise reseller “and the largest user of the product. This will represent the latest advancement of our firm’s investment in AI that will expand our technology ecosystem, bring GenAI deeper into our enterprise, And enable us to scale AI capabilities across businesses to help drive accelerated impact for clients.”
Users benefit, according to Stein, through enhanced data insights, efficient resource allocation, personalized solutions and accelerated problem solving.
Meanwhile, PwC is actively engaged in GenAI discussions with 950 of the firm’s top 1,000 U.S. consulting clients, and is exploring implications of AI on the audit side. PwC has identified more than 3,000 internal GenAI use cases, and more than 95% of U.S. staff have volunteered time to learn My AI, a chatbot for developing AI research and communication skills.
“Corporate IT Spending Isn’t Reflecting the AI Boom,” read a June 26 Yahoo Finance headline. The article cited the contention of Guggenheim software industry analyst John DiFucci that the Big Techs are in a “build it and they will come” mode, scaling data centers and developing LLMs. That “hasn’t made its way to software . . . partly because of cost, and it seems partly because companies haven’t yet figured out what AI is useful for.”
In view of what DiFucci termed a “challenging” IT spending environment, he wrote:
“Most of the spending on AI is being done by AI companies and public cloud companies preparing to run AI workloads for AI companies. That doesn’t mean that we won’t see that shift at some point, when corporations begin to purchase co-pilots and other forms of AI en masse, or start to build their own LLMs as the cost to build and train them continues to decline. But that doesn’t seem to be the case right now.”
Brent Orrell of AEI
In The Hype Around ‘Operational Efficiency’, American Enterprise Institute senior fellow Brent Orrell writes, ”AI integration is the hot new trend – and for good reason.” Sectors including software, healthcare, financial services and professional services “are heavily ‘exposed’ to AI, and those working in them should expect new pressures to adapt to AI-infused systems to speed and improve workflows. Some may also need to transition to new jobs and industries.”
“Yet many organizations are still in the early stages of implementing AI, with clear investment strategies yet to be developed,” Orrell says. “AI may be here, but specific outcomes for workers are still largely unknown.”
MIT Professor Daron Acemoglu, author of Project Syndicate article Don’t Believe the AI Hype, pushes back on the high expectations. He differentiates “easy” tasks that AI can assist in the near term from “hard,” more complex decisioning that will be required to make a measurable impact on GDP.
Along similar lines, Mohammed Hossein Jarrahi, professor in the Information Sciences Center, School of Information and Library Sciences, University of North Carolina, says, “In the short term, AI investments will enhance efficiency and implement cost-cutting measures. They can also raise workers’ productivity by, for example, automating mundane tasks.”
But, Jarrahi adds, too much emphasis on cost-cutting or headcount reduction is shortsighted.
“Companies that succeed with AI are those that truly understand its nature – not just to reduce costs, but to systematically upskill their workforce, focus on strategic partnerships and implement responsible AI principles,” he says. “This strategic approach will be about human-AI symbiosis and goes beyond automation and efficiency gains, aiming to augment the workforce to be more effective.”
Business leaders want to move quickly but are realistic about immediate results.
According a Boston Consulting Group survey of C-suite executives, while almost all “now rank AI and GenAI as a top-three tech priority for 2024, 66% of leaders are ambivalent or dissatisfied with their progress on AI and GenAI – and only 6% have begun upskilling in a meaningful way.” Ninety percent are waiting for AI to move beyond hype, or are only pursuing limited experimentation with the technology.
Challenges cited by those who were dissatisfied with progress on AI and GenAI included a shortage of talent and skills (62%), unclear investment priorities (47%) and absence of a strategy for responsible AI (42%).
Still, 71% were planning to increase their companies’ tech investments in 2024, up from 60% in 2023, and 85% were to increase their spending on AI and GenAI this year.
Source: BCG survey
As McKinsey put it: “Excitement over this technology is palpable, and early pilots are compelling. But a full realization of the technology’s benefits will take time, and leaders in business and society still have considerable challenges to address. These include managing the risks inherent in generative AI, determining what new skills and capabilities the workforce will need, and rethinking core business processes such as retraining and developing new skills.”
Banks have “rightly focused on productivity” in GenAI pilots, “due to the broader pressure on banking economics,” McKinsey stated. It added that “the technology could greatly alter how some jobs are done and how customers interact with banks. It might even lead to entirely new business models.”
“On top of facing tremendous geopolitical risk, financial institutions remain under extraordinary micro- and macroeconomic pressures,” SAS’s Bradley observed. “The rising interest rate environment has created a level of unease and uncertainty that firms hadn’t seen or experienced in more than two decades.”
Bank failures last year exposed the perils of insufficient risk-management rigor, Bradley said, adding that in the wake of Silicon Valley Bank’s collapse, a survey found 80% of firms were considering significant improvements to their asset-and-liability management capabilities.
“Among the most common conversations I'm having with executives these days is around the need for agility in their technology,” Bradley said. “Unfortunately, when I speak with their IT counterparts, I hear they’re buckling under the weight of legacy technology investments.”
Continually layering-on ad hoc or piecemeal solutions is unsustainable, making the necessary agility that much harder to achieve. SAS’s approach is to enable “a more holistic approach to customer decisions – across fraud and financial crimes, across risk management through initiatives like integrated balance sheet management, and even marketing and customer experience. We help organizations streamline and consolidate onto a more consistent architecture and ultimately help them reduce the complexity of their IT environments as they modernize.
“Delivering modularized solutions across the risk, fraud and compliance spectrum on a single, cloud-native data and AI architecture is the very definition of integration and efficiency.”