Agentic AI systems capable of running scams entirely without human intervention could transform financial fraud from a persistent problem into an industrialized crisis, with costs potentially falling by 90% and successful fraud activity more than doubling. Yet financial institutions still have a relatively narrow but valuable window to strengthen their defenses before criminals gain access to these autonomous capabilities, according to analysis from Boston Consulting Group.

The threat is not hypothetical. Agentic capabilities, which allow AI systems to plan and execute complex tasks across multiple steps, have improved tenfold per year since 2024. If this trajectory continues, within one to two years these models will be capable of running day- or week-long scams autonomously, democratizing access to sophisticated fraud operations and potentially drawing thousands of new scammers into the criminal ecosystem.

How Could Agentic AI Industrialize Financial Crime?

Today, scams and fraud already cost consumers and businesses around $442 billion annually, according to the Global Anti-Scam Alliance. But the current landscape relies on human orchestration. Agentic systems would change that fundamentally by automating the entire fraud value chain, from synthetic identity creation and voice cloning to fake document generation and real-time social engineering interactions.

The economic incentive is staggering. When the cost of running scams drops by 90% or more, individual scammers and small teams could generate sophisticated, multi-step fraud campaigns that previously required organized criminal networks. This cost collapse would enable adversaries to launch far larger volumes of attacks, test a broader range of tactics against bank defenses, and adapt rapidly when those defenses evolve.

"What makes agentic systems so concerning is that they could automate the entire fraud value chain. As the cost of running scams and fraud falls dramatically, cybercriminals will be able to scale their operations like never before. As a result, we could see successful scam and fraud activity more than double, with profound emotional and financial costs to those who are targeted," said Othman Omary, Managing Director and Partner and Africa Financial Institutions Node Head at Boston Consulting Group.

Othman Omary, Managing Director and Partner, Boston Consulting Group

The stakes are particularly acute in Africa, where mobile-first banking and digital payments are accelerating financial inclusion at scale. Recent data from TransUnion's H1 2026 Update shows that while suspected digital fraud rates declined in 2025, criminal activity has become more organized and targeted, with greater exploitation of digital identities. In Kenya, median consumer fraud losses reached $839, while South Africa recorded unusually high suspected fraud rates in login attempts.

Why Do Banks Still Have Time to Prepare?

The critical advantage banks hold is timing. Frontier AI models, built by leading labs with sophisticated safeguards against misuse, currently maintain a lead over open-source models. However, open-source models typically catch up with leading capabilities within six to 12 months. This means scammers will likely gain access to agentic models capable of end-to-end fraud within the next couple of years, giving banks a relatively narrow but real window to act.

Leading financial institutions are already demonstrating that proactive investment can materially strengthen resilience. Banks that have worked with BCG to enhance fraud-management capabilities by combining advanced analytics, behavioral monitoring, and AI-enabled intervention tools have improved their ability to identify high-risk activity earlier, reduced the burden on fraud operations teams, and accelerated response times to emerging threats.

How Banks Can Build Defenses Against Agentic Fraud

• Increase Threat Monitoring: As agentic AI developments change the threat landscape at a faster pace, banks need to gather richer behavioral and transactional data to train and continuously retrain detection models. This includes tracking capabilities of open-source, uncensored AI models, building adversarial agents to test a bank's own detection systems, and creating fake customers as scam targets to extract scammer information.
• Leverage the Frontier Model Advantage: Banks should use the lead that frontier models have over open-source models to continuously improve scam and fraud prevention. Key steps include building a strong internal AI team with early access to the latest models and developing rapid approval pathways to get new capabilities into production quickly.
• Build a Responsive, Scalable Operating Model: Banks will need an operating model that can effectively scale up to deal with volume spikes and respond within hours, not weeks. This includes dynamically increasing controls on high-risk payments, real-time analysis of calls to detect new scam patterns, and automating intervention and recovery steps so they become scalable.
• Strengthen Ecosystem Collaboration: As fraud and scam networks become more connected and adaptive, financial institutions will need to work more closely with payment providers, platforms, telcos, social media companies, regulators, and law enforcement to identify and respond to emerging threats.

"The most effective time to act is before operational pressure starts to build. Banks that invest now in strengthening their fraud prevention capabilities will be far better positioned than those forced to respond reactively. That means developing models that not only detect suspicious activity, but can also intervene in real time, freeze transactions when necessary, help customers remove themselves from high-risk situations quickly, and preserve trust when fraud incidents occur," explained Hakim Hamane, Managing Director and Head of BCG Platinion.

Hakim Hamane, Managing Director and Head of BCG Platinion

The broader implication is clear: the financial system faces a more persistent, "always-on" threat environment that requires a fundamentally different defense model than what exists today. Banks that act now to build adaptive, AI-enabled defenses and redesign their operating models for scale will be far better positioned to weather the agentic AI era than those that wait for the crisis to force their hand.