Banks embrace AI coworkers, redefining roles and revenue

By early 2026, the banking industry was undergoing a fundamental shift from AI as mere decision-support tools to autonomous AI agents capable of executing tasks and managing outcomes independently. This agentic transformation redefined operational workflows, moving banks from systems that simply aid human decisions to those that actively execute processes under human oversight, effectively embedding AI as governed digital labor within core execution models. Institutions embracing this shift gained structural advantages including enhanced revenue growth, cost optimization, operational resilience, and accelerated decision-making.

JPMorgan Chase emerged as a pioneer in deploying powerful, long-running autonomous AI agents in 2026, marking a transition from short-lived AI tools to digital coworkers capable of managing complex workflows for hours at a time. Derek Waldron, the bank’s chief analytics officer, described these AI agents as evolving into 'team managers' that can delegate tasks and maintain intellectual coherence over extended periods, signaling that security and governance challenges previously hindering enterprise AI adoption were nearing resolution.

Early AI deployments at JPMorgan have already yielded tangible business benefits, with private banking AI tools contributing to a 20% increase in gross sales and enabling potential client coverage expansion by 50%. Reflecting a strategic pivot, JPMorgan shifted its AI focus from cost-cutting to driving revenue growth and sustainable competitive advantage, while also reevaluating build-versus-buy decisions for AI capabilities to maximize impact.

By mid-2026, major Wall Street banks including Morgan Stanley, UBS, BNY Mellon, JPMorgan, and Goldman Sachs had moved beyond experimental AI chatbots to integrating autonomous AI agents as digital coworkers with formal roles, including logins, human managers, and performance reviews. This integration has delivered measurable performance improvements, particularly in investment management, where JPMorgan’s AI-powered agents outperformed traditional portfolios on a risk-adjusted basis, underscoring the practical value and operational maturity of autonomous AI in banking.

By mid-2026, banks like Merrill Wealth Management and Bank of America Private Bank demonstrated how agentic AI can dramatically enhance operational efficiency by automating labor-intensive tasks such as client meeting preparation and mortgage pre-adjudication, compressing hours of work into minutes. This focus on accelerating repetitive, measurable workflows rather than autonomous decision-making highlights AI’s role as a force multiplier within internal banking operations, enabling employees to redirect efforts toward higher-value activities.

Financial crime investigation has emerged as a prime domain for agentic AI adoption, with institutions like First National Bank of Omaha cutting investigation times by 50% through AI-assisted alert triage and evidence gathering. Nasdaq Verafin’s AI agents, deployed across over 650 banks, have achieved up to a 90% reduction in sanctions alert review workload, underscoring how AI agents are transitioning from pilots to production environments where they handle high-volume, repetitive tasks while human analysts retain final judgment, addressing the escalating complexity of regulatory compliance and illicit activity.

Banks are expanding agentic AI beyond backend automation into customer engagement and core operational roles, as seen with Slash’s 'Twin' AI Chief of Staff enabling natural language dashboard interactions, and Wall Street giants like UBS, Morgan Stanley, and BNY Mellon integrating AI agents as digital coworkers with managerial oversight. These agents not only generate actionable alerts and execute trades upon approval but also dynamically manage portfolios and client interactions, signaling a shift toward AI agents as embedded collaborators rather than mere tools.

Innovations in agentic AI are driving autonomous banking capabilities worldwide, exemplified by Pakistan’s Wateen, which deploys AI agents that reason and decide independently across onboarding, KYC, fraud detection, and compliance monitoring. By integrating AI reasoning with workflow orchestration tools like n8n, Wateen enables seamless, auditable actions within secure on-premise infrastructures, ensuring regulatory compliance and continuous risk management throughout the customer lifecycle. Similarly, banks adopting AI agents for incident resolution report cutting response times from over two hours to under 30 minutes, with nearly 40% of incidents managed without human intervention, though challenges remain in guaranteeing explainability and reversibility to satisfy stringent oversight requirements.

Despite significant AI advancements in MSME lending, India faces challenges in expanding credit access to new borrowers, as evidenced by the decline in first-time MSME loan recipients from 52% to 42% over three years. Indian banks like Absa Bank emphasize AI-driven hyper-personalisation to better understand customer needs and tailor credit products, yet human oversight remains crucial to mitigate algorithmic bias and ensure fairness, especially for informal traders and freelancers.

Indian Bank’s launch of a unified digital platform for MSME loans up to Rs 5 crore exemplifies how automation and digital identifiers such as PAN, GST, and Udyam Registration are streamlining loan processing and underwriting, enabling in-principle approvals within 30 minutes. Backed by a robust MSME portfolio exceeding Rs 1.10 lakh crore and 85 processing centers, the bank’s strategic focus on growth hubs like Delhi-NCR underscores its commitment to expanding credit access through scalable digital innovation.

At FIBAC 2026, major Indian banks including SBI, HDFC Bank, and Bank of Baroda highlighted the transition of Agentic AI from pilots to core operations, enhancing lending efficiency and customer service while emphasizing the critical role of high-quality training data and robust governance frameworks. RBI Governor Sanjay Malhotra stressed board-level accountability and human-in-the-loop oversight to prevent systemic risks, as banks balance AI-driven gains with concerns over cybersecurity, data privacy, and potential bias in credit risk models.

State Bank of India’s AI strategy showcases large-scale underwriting of Rs 1 lakh crore in MSME loans up to Rs 5 crore within FY26, leveraging AI—including Large Language Models—to automate cheque processing and free relationship managers from routine tasks. This comprehensive AI integration spans credit underwriting, portfolio management, and fraud risk, with continuous model training and control units ensuring accuracy, while early warning systems analyze sector-specific data to preempt delinquencies and strengthen MSME risk management.

By mid-2026, Indian banks have moved Agentic AI from experimental pilots to integral components of their operations, underscoring the critical role of governance frameworks that embed board-level accountability and human-in-the-loop oversight. RBI Governor Sanjay Malhotra stressed that banks must proactively direct their AI strategies rather than passively allowing technology to dictate sectoral shifts, highlighting the necessity of maintaining human judgment alongside automation to ensure responsible AI deployment.

A cornerstone of responsible AI adoption in banking lies in rigorous data governance, with institutions like HDFC Bank prioritizing the quality and precise labeling of training datasets to avoid pitfalls such as outdated information skewing credit decisions. Additionally, banks are advancing customer data tokenization techniques to bolster privacy protections, reflecting a dual focus on transparency and security as essential pillars in mitigating AI-related risks.

Navigating the tension between rapid AI-driven automation and stringent risk management remains a formidable challenge, as regulators like the RBI caution against systemic vulnerabilities including AI hallucinations, biased credit models, and escalating cyber threats. In response, banks are investing heavily in upgrading data infrastructure and reskilling employees, striving to harmonize innovation with robust cybersecurity and compliance frameworks to safeguard both institutional integrity and customer trust.

In August 2026, Suryoday Small Finance Bank marked a strategic pivot toward internal AI capability building by partnering with Kyndryl to launch an Agentic AI Center of Excellence. This collaboration leverages Kyndryl’s consult-led approach and proprietary Agentic AI Framework to embed intelligence across critical customer journeys and operational processes, accelerating onboarding, enhancing responsiveness, and enabling staff to focus on higher-value tasks. As Baskar Babu R, MD and CEO of Suryoday Bank, emphasized, this initiative not only modernizes operations but also fosters a resilient and governed innovation ecosystem aligned with long-term growth and evolving customer expectations.

Banks are investing heavily in data infrastructure and workforce reskilling to balance rapid AI adoption with the need for robust cybersecurity, compliance, and institutional trust.