Bank of America's chief technology and information officer said one of the most common mistakes companies make with artificial intelligence is reaching for it first — even as his own bank prepares to double its AI budget next year.

The remarks offer a snapshot of how two pillars of global finance are deciding where AI earns its keep — and where simpler tools do the job better.

"One of the biggest mistakes we see us and others doing is rush to AI as a solution," Hari Gopalkrishnan said at Fortune's AIQ Summit in New York, "when deterministic models do a plenty good job."

Gopalkrishnan appeared on the panel alongside Sally Moore, S&P Global's chief client officer and co-head of Kensho Data & Platforms. Fortune Editorial Director Andrew Nusca moderated the discussion.

Bank of America: Simple Tools First

Gopalkrishnan said the bank begins with what clients need and builds a "process inventory" of the steps behind their requests. It frequently decides against AI — "plenty of times," he said — concluding that a mobile app or a real-time decision rule can be the better answer.

Every AI project at the bank also passes through a review covering 16 "pillars" of risk, including privacy, bias, workforce impact and intellectual property. "We're not going to implement a chatbot that only answers to certain accents," he said.

The bank has used AI for more than a decade, beginning with fraud models, he said. Its Erica virtual assistant has handled 3.6 billion transactions, according to Gopalkrishnan, who added that without it the bank would need 11,000 more people to answer customer calls. A March bank press release counted Erica's client interactions at more than 3.2 billion.

That caution coexists with heavy spending CEO Brian Moynihan said in September that roughly 140 AI applications cost the bank $400 million and generate $800 million in benefit, and that the AI expense budget will double next year. The doubling of budgets alongside that methodical review captures the tension the panel kept returning to: invest heavily, but only where the technology demonstrably works.

The spending is routed carefully. Gopalkrishnan said the bank is model-agnostic: an orchestration layer, called Orchestra, sends simple classification tasks to approved open-weight models running on the bank's own GPUs, while harder reasoning work goes to proprietary models. He said the setup also helps control token costs.

In wealth management, he said, advisors can now prepare for client meetings in "seconds and minutes," work that used to take days and weeks.

On autonomous agents, the bank is in no hurry. "There is so much juice to be squeezed right now with assistive agents that are actually working with humans in the loop," Gopalkrishnan said. The bank will go further as control infrastructure improves.

He also addressed security, noting that AI models are getting better at finding software vulnerabilities, which makes patching and secure development essential whether or not a company deploys AI. The stakes extend beyond any single institution, he argued: if a small bank somewhere has a problem, people lose faith in the financial system.

S&P Global: Data as the Currency

Moore said her 160-year-old company is repositioning itself — aggressively and carefully. S&P Global is the world's largest credit rating agency and one of the largest index providers, and much of its financial data feeds regulated workflows, she said.

"Data is the currency within AI," Moore said, adding that clients need accuracy, citations, and auditability and traceability back to the source. In a business whose data feeds regulated workflows, that auditability is the condition for using AI at all.

She credited an early bet: S&P Global acquired the AI company Kensho in 2018, a move that has "given us an advantage," she said, explaining that the company has since placed Kensho at the center of the business. On July 6, it split its Market Intelligence division into two units. The first, Kensho Data & Platforms, pairs "Kensho Data" — the client-facing data and AI delivery layer — with a Platforms group housing Capital IQ, Ratings Direct, Visible Alpha and With Intelligence. The second unit is Enterprise Solutions. CEO Martina Cheung said the changes should support revenue growth and better margins. The reorganization gives those stated goals a concrete test of whether an AI-centered data strategy pays off.

About two years ago, S&P Global also created a chief client office, which Moore leads, to work more closely with clients. It includes a labs group and what the company calls forward-deployed experts, who work alongside clients on AI.

S&P Global serves 60,000 clients at different stages of AI adoption, Moore said. The company works with frontier AI labs, puts its data into large language models and productivity tools, and now builds its own agents. Use cases run from broad tasks, such as bankers preparing for meetings and pitch books, to specialized agents built for a single job.

She offered one example. A tier-one bank with 8,000 bankers was combining S&P content sets within its own platform. S&P helped bring the project to production "six times quicker," she said, and accuracy rose from about 60% when the bank started to 98% afterward. She did not name the bank, and the figures are S&P's own.

The company has also built a "credit memo builder" agent that keeps humans in the loop and relies on confidence in the underlying data.

Like Gopalkrishnan, Moore emphasized using the right tool for the job. S&P Global's deterministic option for language models is an API, which does not run up heavy token costs, she said, while more complex "adaptive data retrieval at scale" costs more. Because clients pay for data, technology and people efficiency from different budgets, S&P discusses a range of outcomes with them.

Where They Part Ways

Asked whether he ever chooses against AI, Gopalkrishnan said the simplest answer is often the best one. "AI is not always the right answer," he said.

Moore agreed, then took the point further. "I think Hari said it really well. It's not around the right tool. It's a little bit about reinvention," she said. "What am I trying to solve for here? And where can this technology take me?"

Inside S&P Global, that has meant a central transformation office that brings together technologists and, more recently, data operations. Both companies have left measurable markers for the months ahead: the bank's doubled AI budget takes effect next year, and S&P Global's reorganized divisions will be judged against Cheung's growth and margin targets.

In the closing lightning round, Moore pointed to "embedded intelligence," which she said "creates an opportunity to go beyond the clients that you serve today."

Gopalkrishnan had the last word: "I think anticipating your client needs and serving them where they are will differentiate you in otherwise commoditized space."

Fortune journalists used generative AI as a research tool for this story, and an editor verified the accuracy of the information before publishing. This story was originally featured on Fortune.com.