Artificial general intelligence and superintelligence “are going to change humanity and the human condition.” So says Demis Hassabis, Nobel Prize winner and co-founder and CEO of Google DeepMind.
While occupying that theoretical, futuristic dimension, AI is also a real, everyday business proposition, raising more immediate if not urgent questions about the nature of employment and eliciting pragmatic responses from major institutions and their leaders.
Goldman Sachs chief executive officer David Solomon told Bloomberg that even as AI is being rapidly and widely deployed, he believes it will more likely improve workflows and productivity than displace a lot of jobs.
Citadel founder and CEO Kenneth Griffin, on a recent Goldman Sachs podcast discussing AI’s “human capital implications,” stressed opportunity and productivity: “There’s no reduction to headcount at Citadel . . . I have incredibly talented people. We have just a huge swath of problems that we’re trying to attack and go after. I will take every single productivity gain I can get because with the talented people we have, we just have more to go after.”
JPM’s Dimon: “Better job for the customers.”
During JPMorgan Chase & Co.’s July 14 earnings call, chairman and CEO Jamie Dimon said that AI has enabled job cuts of up to 40% in some parts of the bank but with limited impact on overall operating expenses. "You don’t uniquely benefit from AI,” he said. “In a competitive, capitalist world, we all will use AI to do a better job for the customers. We can’t just say, ‘Oh, it's going to increase our margins’ . . . If that were true, our margins would be 80% today because of computerization over the last 20 years.”
These remarks may reflect a consensus regarding the current state. But there is, in fact, no shortage of observations on AI implementations and implications from CEOs and their management teams, from consulting firms and think tanks, and from regulatory officials concerned about operational risks and systemic resilience.
Hassabis weighed in on regulation in a July 14 X post, A Framework for Frontier AI and the Dawning of a New Age. The DeepMind visionary suggested forming a “standards body modeled on a federally overseen public-private partnership or self-regulatory organization, much like the Financial Industry Regulatory Authority (FINRA), with a board that includes independent leading technical experts and open-source representatives.”
That proposal has gotten positive feedback from tech leaders including Microsoft CEO Satya Nadella (Microsoft AI is headed by DeepMind co-founder Mustafa Suleyman) and is reportedly under consideration by the Trump administration.
The World Economic Forum (WEF), for one, has an AI in Financial Services initiative and, with Accenture, produced a playbook offering “practical guidance for organizations at every stage of their AI journey” and “outlining the frameworks needed to scale AI responsibly and securely.”
The 2026 Global AI in Financial Services Report out of the Cambridge Centre for Alternative Finance (CCAF), University of Cambridge – jointly with the WEF, Bank for International Settlements, International Monetary Fund and other collaborators – cited risks and challenges in such areas as skills and data quality, agentic AI “scaling cyber risk beyond the reach of manual oversight”, and an “explainability-expectation gap.”
Among the CCAF headline findings: “Productivity gains are visible, enterprise value is harder to prove”; and “regulators are generally optimistic about AI’s role in achieving their objectives by 2030.”
Source: CCAF Global AI in Financial Services Report.
At its June 1 plenary, the multinational Financial Stability Board (FSB) “highlighted two new developments that have further complicated the risk landscape”: potential knock-on market effects from the Middle East conflict; and “the unveiling of powerful frontier AI models has concerned regulators and market participants. Such models may sharply increase cyber risks, and patching efforts are important, but may add to problems if rushed or poorly executed.”
The FSB on June 10 released a consultation report, Sound Practices for Responsible Adoption of Artificial Intelligence, “to help the board and senior management of financial institutions as they consider business strategy, technology adoption, and risk management in an increasingly AI-enabled environment.”
Michelle W. Bowman, the Federal Reserve Board’s vice chair for supervision and chair of the FSB’s Standing Committee on Supervisory and Regulatory Cooperation, said on July 7 that Sound Practices and its case-study examples were informed by the Fed’s observing “a noticeable increase in the use of AI by banks of all sizes, and we have seen a variety of use cases. Our focus has been on supporting institutions that want to innovate responsibly by leveraging AI tools in their operations.”
As she has done in other supervisory contexts, Bowman underscored “materiality . . . We emphasized that lower-risk uses of AI should receive a lighter supervisory and regulatory touch”; and proportionality: “What works or is a consideration for larger institutions using AI in complex applications is not appropriate for smaller institutions with less complex AI uses. As I noted, our focus is on promoting innovation at financial institutions of all sizes, not just the largest ones. This report provides clear guidance to all institutions, including smaller institutions.”
“Artificial intelligence is reshaping how financial firms price risk, allocate credit, and respond to stress. It is increasingly embedded in the decision‑making architecture of the financial system,” the International Monetary Fund’s Tobias Adrian wrote in a July 23 IMF Blog, How Central Banks Can Contain Financial Stability Risks as AI Accelerates Change.
It has become “a financial‑stability issue that cuts across markets, institutions, financial infrastructures, and supervision,” said Adrian, financial counsellor and director of the fund’s Monetary and Capital Markets Department. “In an AI‑enabled financial system, stability will depend less on any single model and more on the institutions, incentives, and safeguards that govern their use . . . If policymakers act early and collectively, AI can reinforce global financial resilience. If they do not, future instability may be faster, more correlated, and harder to manage than past episodes.”
Between survey data and strategic advice, the consulting sector offers a variety of perspectives on AI-related themes and management priorities.
“Banks must develop a new and increased velocity of execution, to match the speed of AI development,” McKinsey & Co. asserted in a preview of its October 2026 Global Banking Annual Review. Banks got by in earlier technology waves because their most profitable customer segments were not fast adopters, but “this time there will be no grace period,” the firm warned.
“ChatGPT launched in November 2022; by 2024, fully 45% of U.S. working-age adults were using gen[erative] AI, rising to 55% by 2025. The usual generational gaps are not as prominent as in the past; the gap in usage between young people and older adopters is small.
“Moreover,” McKinsey went on, “people are also entrusting the technology with vastly more complex tasks – which was both not technologically possible with other technologies, and, because of the earlier tech’s limitations, was beyond the limits of trust for consumers.”
Source: World Economic Forum, The AI Playbook for Financial Services.
Boston Consulting Group titled a June report, “Future of Finance 2026: Time to Shift Gears?” – a loaded question.
"Financial institutions have had an exceptional year, but the market is telling them something important: Past performance is not enough,” Saurabh Tripathi, global leader of BCG’s Financial Institutions practice and a co-author of the report, said in a press release. “The P/E discount reflects genuine investor skepticism about whether banks can deliver sustained, compounding growth. Institutions that act now to redesign their operating models, redeploy capital into growth, and build AI into their strategic core have a real opportunity to close that gap. Those that don't will find that it widens.”
Saurabh Tripathi of BCG
BCG concluded that despite years of significant technology investment, operating expenses relative to assets have improved only marginally; financial industry headcount has grown approximately 2% annually over the past three years; and digitization has layered technology onto existing processes rather than fundamentally reimagining them.
Vivek Singh, banking head, Capgemini Research Institute, said that in a global corporate and investment banking survey, 82% responded that AI-driven products had not yet increased revenue or delivered expected cost savings.
Meanwhile, BCG said, financial institutions plan to invest 2% of revenue in AI this year, with only the tech industry spending more.
“The productivity problem in banking is structural, not cyclical, and incremental digitization has not solved it,” said BCG managing director and partner Andreas Biffar. “The institutions that are pulling ahead are rebuilding how they operate from the ground up, with AI at the center. The gains are already measurable, and the gap between leaders and laggards is widening faster than was anticipated one or two years ago.”
Competitive advantage will come through “treating AI as a core strategic capability, applying it to compliance, customer engagement, underwriting and account management,” contends Shawn Ellis, managing partner at investment firm Distributed Ventures. Wide-scale AI adoption must be underpinned by trustworthy data and strong governance strategies.
“Banks invest in sophisticated tools, then let them collect dust because they lack the governance structure to deploy them confidently,” notes Terry Mendez, CEO of Safe Harbor Financial. That’s a “misalignment between innovation appetite and governance maturity.”
Being a first mover is not a sole determinant, according to Mark Blake, financial services practice lead at master data management provider Stibo Systems. Banks are already experimenting with AI across customer service, onboarding and anti-fraud. The real question is who can embed it into core decision-making, credit analysis, risk management and financial crime protection in a way that is consistent and scalable.
“That’s where data becomes the differentiator,” Blake explains. “If the underlying data isn’t trusted, AI doesn’t solve the problem. It amplifies it. The banks that get their data foundations right are the ones that turn AI into something repeatable, not just experimental.”
Top three AI focus areas by market segment, from WEF’s The AI Playbook for Financial Services.
A SAS-IDC AI Impact Report examines “the relationship between the perceived trust in AI systems and their actual trustworthiness, illustrating the ‘trust dilemma.’ This misalignment, evident across all regions, represents a critical barrier to effective AI adoption.”
The dilemma is said to affect 46% of organizations worldwide. “It is slightly more pronounced in Asia-Pacific and North America, where 47% of organizations face misalignment between trust in AI and actual system trustworthiness.”
“Most banks’ foundational readiness is nowhere near where it needs to be,” comments Stu Bradley, SAS senior vice president of Risk, Fraud and Compliance Solutions. “Roughly nine in 10 banks have yet to fully align trust with proof, and about one in five are still running on siloed data. Closing the gap between AI ambition and AI readiness should be a top-down priority for all banks.”
Through Distributed Ventures’ investment portfolio, Ellis gets what he calls a “direct line of sight” into how well financial institutions are implementing AI. It boils down to a few key questions:
Shawn Ellis of Distributed Ventures
Does the bank have genuine executive-level sponsorship for AI initiatives, or is it being driven by a single champion with no organizational backing? Does the institution have a clean data architecture that allows new tools to integrate without years of remediation? Is leadership tracking measurable outcomes – faster processing times, improved underwriting accuracy, better retention or NPS (net promoter score, a measure of customer loyalty)?
The banks generating real ROI in the early stages of enterprise AI are moving beyond pilots and are operationalizing it across core workflows, Ellis says. Although there is no single formula for success, there are reliable indicators. They include data maturity, including the governance aspect; the pilot-to-production pathway – not contributing to the upwards of 50% abandonment rate on AI initiatives; and vendor ecosystems characterized by productive partnerships with fintechs and third-party solutions.
“Over the next five to 10 years, AI will fundamentally reshape how banks operate,” Ellis maintains. With adoption of agentic AI, it will be increasingly embedded in non-financial platforms and change what traditional banking looks like. On the workforce level, fewer people will be performing routine processing, and more will be managing, validating and governing the AI systems that are.
Oversight of AI models and navigation of evolving regulatory standards become key risk management responsibilities. Safe Harbor Financial’s Terry Mendez envisions “a small army of agents that are able to catch problems and identify opportunities earlier than ever before.” Risk managers will be translators between the technologists building these systems and the executives accountable for them.