Brazil has long been one of the world’s most advanced markets for digital financial services. Instant payments, mobile banking, biometric authentication, and digital identity have created fast, convenient experiences at enormous scale. That same maturity, however, has also made the market a proving ground for increasingly sophisticated fraud.
Generative AI is accelerating that dynamic. It is not simply introducing new attack techniques; it is changing the economics of fraud by reducing the cost, expertise, and time required to create convincing attacks at scale.
That reality shaped a recent Aware event in Brazil that brought together leaders from banking, fraud prevention, cybersecurity, product security, digital identity, and technology. The discussion ranged from deepfakes and injection attacks to biometric authentication, orchestration, risk-based security, and the future of identity infrastructure.
Across those conversations, one idea stood out: the next phase of fraud prevention will depend less on any single technology and more on an organization’s ability to adapt.
AI is Shifting the Advantage Toward Scale
Fraud has always evolved alongside technology, but AI is accelerating the pace of that evolution. Attackers can increasingly automate social engineering, manipulate identity evidence, create synthetic content, impersonate legitimate users, and test new techniques with far less effort than before.
Deepfakes are one of the most visible manifestations of this shift, but the broader concern is industrialization. AI allows sophisticated techniques to be repeated, refined, and deployed at higher volume, creating an environment in which defenders may face more attacks, more variations of those attacks, and less time to respond.
That changes the strategic question for financial institutions. It is no longer enough to ask whether a particular fraud tool performs well against a known threat. Organizations also need to ask whether their broader identity architecture can adjust when the threat changes.
This is especially important in Brazil, where the scale and maturity of digital financial services mean that both innovation and risk tend to move quickly. The organizations that respond most effectively will be those that build adaptability into their identity strategy rather than treating it as a future enhancement.
Identity Security Must Extend Beyond the Point of Entry
Many identity programs have historically concentrated their strongest controls around onboarding and authentication. Those moments remain important, but they no longer define the entire risk surface.
Fraud can emerge during account recovery, device changes, transactions, profile updates, or other points in the customer lifecycle. A user who was legitimately verified at enrollment can still be compromised later. A legitimate customer can be socially engineered. Manipulated identity evidence can enter a process after the initial account has already been established.
The implication is that identity should increasingly be treated as a continuous risk decision rather than a one-time verification event.
That does not mean introducing more friction everywhere. In fact, the opposite should be the goal. Organizations need the ability to adjust assurance based on context, applying stronger controls when risk rises while allowing legitimate, lower-risk interactions to remain simple.
The strategic value comes from making security proportional. Strong identity programs should be capable of increasing confidence when needed without forcing every customer through the same experience.
Biometrics are Foundational, but Context is What Makes Them Powerful
Biometrics remain central to modern identity security because they provide a strong signal connecting a digital interaction to a human being. Yet one of the clearest conclusions from the event was that biometrics are most valuable when they operate as part of a broader decision framework.
No individual modality or algorithm can address deepfakes, presentation attacks, injection attacks, synthetic identities, compromised devices, and transactional fraud simultaneously. Each threat introduces different signals and different forms of risk.
The stronger approach is defense in depth: biometric matching, liveness detection, document verification, device intelligence, behavioral data, transaction risk, and other contextual signals working together.
This is not simply a technology architecture decision. It is a decision about how confidence is established.
Biometrics remain a critical layer of identity security, providing organizations with a strong way to establish confidence in who is interacting with their systems. Within a defense-in-depth strategy, biometrics can work alongside liveness detection, device intelligence, behavioral signals, transaction risk, and other contextual data to create multiple layers of protection. By combining these complementary signals, financial institutions can make stronger, risk-based decisions while adapting the level of assurance to the transaction, customer journey, and threat environment.
Orchestration is Becoming a Strategic Capability
As financial institutions introduce more identity and fraud technologies, complexity can quickly become its own source of risk. Different tools may be owned by different teams, integrated at different times, and optimized around different use cases. The technical challenge of connecting those systems is relatively straightforward compared with the strategic challenge of coordinating them.
Organizations need to determine which technologies should be invoked, when additional assurance is necessary, how different signals should influence a decision, and how workflows should change as new risks emerge.
That is why workflow and biometric orchestration are becoming more important. Its value is not simply in linking APIs, but in creating a framework through which organizations can manage technologies, vendors, policies, and decisions as part of a coherent identity strategy.
The strongest orchestration models should also make change easier. Financial institutions need the ability to introduce new technologies, compare providers, change workflows, or replace underperforming capabilities without having to redesign the entire identity stack.
In an environment where fraud techniques are changing rapidly, that flexibility is not just an operational benefit, but also a security capability.
Open Orchestration Protects Strategic Choice
Not all orchestration approaches provide the same level of control. Some platforms offer predefined services, basic vendor switching, or what is essentially load balancing, while others obscure decisions inside closed ecosystems. For financial institutions, that distinction matters.
An open orchestration model should preserve the organization’s ability to choose which technologies it uses, how those technologies interact, and how the identity strategy evolves. It should also provide visibility into performance so that teams can understand whether a particular vendor, model, or workflow is actually delivering the intended outcome.
Vendor independence is therefore not simply a procurement consideration. It is part of resilience.
A rapidly changing threat environment makes it difficult to assume that one provider, model, or approach will remain optimal indefinitely. Organizations need the freedom to adapt based on performance, risk, customer experience, and emerging threats rather than the commercial interests of an orchestration provider.
Adaptability is Becoming the Defining Advantage
No financial institution can predict every fraud technique that will emerge in the next several years. AI makes that challenge even more difficult because it accelerates the rate at which attackers can experiment.
The more durable strategy is to build an identity environment capable of changing with the threat. That means extending protection across the customer journey, applying assurance according to risk, combining complementary signals, maintaining visibility into performance, and preserving the ability to introduce or replace technologies as conditions evolve. All that, while also taking context and transaction risk into account.
The objective should not be to create an identity infrastructure that remains unchanged for years. It should be to create one that can evolve without losing control. As AI continues to lower the cost and increase the scale of sophisticated fraud, that adaptability may become one of the most important advantages financial institutions can build.
Building a more adaptive identity strategy requires both stronger fraud defenses and the flexibility to evolve as threats change. Learn more about how Aware Intelligent Liveness supports trusted identity verification and how the Awareness Platform helps organizations orchestrate biometric technologies, vendors, and workflows with greater control.
