AI Agent Research | Nov 10, 2025

Image: McKinsey and Google Cloud AI in Finance 2025 report covers

Insights from 2025 Google Cloud and McKinsey AI Finance Studies

November 2025 research from McKinsey's The State of AI in 2025 and Google's The ROI of AI in Financial Services reveal that the way banks measure value from AI is changing. Instead of chasing fast gains from early generative tools, the financial sector is moving toward agentic AI platforms that produce steady operational returns and stronger control over risk and cost.

See:  Take Part in the Global AI in Finance 2030 Survey

• 77%of financial institutions report positive ROI from generative AI (about the same as last year), confirming steady value creation rather than hype
• 76% of executives cite productivity gains and 67% improved customer experience, with new AI use cases reaching market in three to six months
• 49% of firms plan to allocate over half of next year’s AI budget to agentic programs, a structural change from models to multi-agent systems
• High performers are over three times more likely to redesign workflows, have direct executive ownership of AI programs, and spend over 20% of digital budgets on AI

The ROI Story Behind the Numbers

Google’s global financial services study found that 77% of firms report positive ROI from generative AI this year, matching last year’s figure. The data point indicates that the return base has stabilized after an initial surge (not stagnation or decline). Among responders reporting higher revenue, 51% saw growth (6%-10%), while 30% achieved gains above 10%. These numbers dipped slightly year over year, pointing to a normalization phase as programs scale.

See:  SEC Chair Backs Crypto Super Apps and AI Finance at OECD

The same study found that:

• 63% of firms saw overall business growth from AI
• 76% cited productivity improvement
• 67% of executives said customer experience improved, even though satisfaction scores were flat which is a sign that front-line automation works well for simple tasks but needs refinement for complex customer issues
• 52% of firms reported an improvement in security (slight decline as more move to specialized agentic systems to handle complex AI threats)

Spending Focus From Models to Agents

Spending priorities have shifted decisively:

• 53% of financial services executives now deploy AI agents
• 40% run more than ten in production
• 49% expect to allocate over half of next year’s AI budget to agentic programs
• Falling model costs have led 79% of institutions to increase AI spending (61% funding these initiatives through new capital instead of repurposed budgets)

See:  Ripple’s $500 million Investment and 2 Years of Expansion

McKinsey’s global findings confirm this rotation:

• 88% of organizations report regular AI use
• About 33% say they have scaled it across the enterprise
• 62% are testing agents while 23% are already scaling them
• Less than 10% have achieved full deployment in any single business function

High performers stand out whose choices correlate directly with both revenue growth and cost reductions.

• 3x more likely to pursue transformation rather than incremental gains
• 3x more likely to have executive-level accountability for AI programs
• 5x more likely to allocate over 20% of digital budgets to AI

Banks Measurable Impact

Benefits are appearing fastest in software engineering, IT, customer service, and marketing. The McKinsey report found that organizations deploying AI across these functions experienced:

See:  How Fintechs Can Build Enterprise AI Agents with Cohere

• Average productivity gains of 15%-30% in service and technology teams
• Cost decreases of up to 12% in customer operations and risk processing
• Deployment cycles reduced to three to six months, down from an average of twelve months in 2023
• Revenue growth between 6%-10% for over half of financial institutions using AI in production
• Improved fraud detection accuracy by up to 20%, especially among banks integrating multi-agent risk models

Organizations combining workflow redesign with governance and human oversight achieve higher and more consistent ROI than those focused only on model performance.

Outlook

According to Google and McKinsey's 2025 AI in finance research reports, AI has entered a new phase. The early boom in generative tools has given way to a steadier phase of operational performance driven by agents. The majority of institutions are already realizing returns, but the next iteration of value capture will come from how well they govern, measure, and scale these systems. Canada’s Opportunity In Efficient Reasoning AI its competitive edge, using discipline, design, and data to turn intelligence into growth.

The National Crowdfunding & Fintech Association (NCFA Canada) is a financial innovation ecosystem that provides education, market intelligence, industry stewardship, networking and funding opportunities and services to thousands of community members and works closely with industry, government, partners and affiliates to create a vibrant and innovative fintech and funding industry in Canada. Decentralized and distributed, NCFA is engaged with global stakeholders and helps incubate projects and investment in fintech, alternative finance, crowdfunding, peer-to-peer finance, payments, digital assets and tokens, artificial intelligence, blockchain, cryptocurrency, regtech, and insurtech sectors. Join Canada's Fintech & Funding Community today FREE! Or become a contributing member and get perks. For more information, please visit: www.ncfacanada.org

Canadian Public Markets Study Seeks Industry Input

Digital Wallet Control Points for Banks and Fintechs

OSFI Flags Frontier AI Risks Across Canadian Finance

FNZ Turnaround Tests La Caisse and Canadian Pension Capital

Senso’s Path From Fintech AI to Zeta Acquisition

National Security Is Now Part of Financial Supervision

How Canadians Pay Is Changing Before Real-Time Rail

Canadians Cut Holiday Spend but Still Pay More for Canadian

Ondo Launches Tokenized Pre IPO Private Markets

Snaplii Puts AI Agent Spending in a Separate Wallet