Cyber ​​risk associated with artificial intelligence (AI) is an issue that deserves careful attention, especially in the context of the global financial system.

In this article, we will explore the Financial Stability Board's (FSB) warning to the G20 regarding the adoption of advanced AI models and the lack of regulatory preparedness in many countries.

Furthermore, we will discuss the financial sector's reliance on a few technology providers and how this can impact market confidence.

Finally, we will address the operational challenges and the need for a safe and responsible release of AI systems to ensure resilience in the financial sector.

FSB Alert on AI and Immediate Cyber ​​Risk

The chairman of the Financial Stability Board, according to the FSB's official website, warned that the advancement of artificial intelligence has already raised cyber risk to an immediate concern for the global financial system , because the technology can amplify the scale, speed, and precision of attacks against banks, insurance companies, and critical infrastructure.

Furthermore, the increasing use of a few technology providers concentrates vulnerabilities and can compromise market confidence, especially when many countries still lack robust systems to control the adoption of advanced models.

In this scenario, AI also facilitates the discovery of hidden flaws, which puts pressure on security teams to fix vulnerabilities much more quickly and under greater operational risk.

The warning becomes even more serious because financial resilience depends on coordinated responses, and any delay can lead to disruptions, data loss, and cascading effects.

As the official emphasized, "The threat is immediate and we cannot underestimate its scope."

Fragility of regulatory systems in G20 countries

Many G20 members still lack adequate regulatory systems because the evolution of AI is advancing faster than the supervisory capacity of regulators.

Furthermore, several countries lack the technical teams, integrated databases, and standardized tests to assess the risks of advanced AI models in the financial sector.

As a result, banks and fintechs may adopt opaque solutions without sufficient control over biases, operational flaws, and misuse of data.

The reliance on a few global suppliers also increases the concentration of risk and makes continuous monitoring more difficult.

In parallel, the fragmentation between national rules creates legal uncertainty and reduces coordination between authorities, which delays responses to cyber incidents and urgent fixes.

"The safe and responsible release of AI requires resilience, preparedness, and constant oversight."

"

• Lack of clear policies
• Insufficient technical infrastructure
• Poor regulatory coordination

Concentration of technology suppliers and risks to market confidence.

The financial sector's reliance on a few technology providers of AI solutions increases its exposure to chain failures, because a disruption, faulty update, or security incident can affect multiple institutions simultaneously.

Furthermore, when banks and fintechs concentrate critical functions in a limited number of providers, they reduce their responsiveness and lose room to negotiate safer standards, which increases operational and regulatory vulnerability.

This scenario worries international authorities, especially because AI can accelerate the discovery of flaws and require faster corrections than traditional structures can deliver.

The higher the concentration, the greater the risk of problems spreading among interconnected institutions.

• Increased systemic risk
• Excessive operational dependence
• Leakage or manipulation of sensitive data

There is also a risk of decreased resilience, as fewer providers become centralized in knowledge, updates, and support.

Therefore, confidence in the market becomes more fragile, as investors and customers begin to perceive that the system's solidity depends on a limited technological link, capable of compromising credibility, continuity of services, and financial stability.

AI accelerates the discovery of vulnerabilities and poses operational challenges.

AI is expanding the ability to identify flaws in financial systems much more quickly because it analyzes large volumes of code, traffic, and anomalous behavior in near real-time.

As a result, vulnerabilities that previously took days or weeks to appear can now be exposed in just a few hours, increasing the pressure for rapid fixes and reducing the scope for extensive testing.

In this scenario, the technology area needs to respond more quickly without compromising stability, since any hasty adjustments can generate side effects in critical integrations, digital channels, and payment operations.

Thus, the gain in detection also creates a new risk: the need for continuous remediation in highly sensitive environments.

Furthermore, the speed of AI exposes significant operational challenges , as teams need to prioritize incidents, validate changes, and maintain service continuity simultaneously.

This requires governance, secure automation, and more robust contingency plans.

As the Financial Stability Board warned,

"The safe and responsible release of AI models is essential to preserving the trust and resilience of the global financial system."

Therefore, the issue is not only about detecting failures earlier, but also about sustaining an equally rapid, coordinated, and resilient response.

Recent incident highlights flaws in AI systems.

A recent episode drew attention by showing that AI models can bypass protective barriers when given instructions designed to induce inappropriate responses.

Reports from the technology sector indicated that an advanced system managed to unexpectedly bypass safeguards, exposing weaknesses in the validation process before production release.

.

This type of failure is concerning because, in the financial environment, AI already supports risk analysis, customer service, and fraud detection.

Thus, any breach expands the attack surface and can compromise sensitive data, automated decisions, and customer trust.

For the financial sector , the impact is immediate, as dependence on a few technology providers creates a ripple effect when a model fails or is manipulated.

Furthermore, AI can also accelerate the discovery of cyber vulnerabilities, requiring faster fixes and stricter controls.

The FSB emphasized that the adoption of advanced models needs to occur with safe and responsible release.

innovation without resilience exposes institutions to operational and reputational risks.

Therefore, banks and regulators must test, monitor, and prepare response plans before incidents become systemic.

Guidelines for the safe and responsible release of AI models.

The Financial Stability Board recommends the safe release of advanced AI models because the financial system depends on trust, operational continuity, and a rapid response to emerging risks.

As warned by the FSB, AI can accelerate the discovery of cyber vulnerabilities and therefore require immediate fixes, increasing pressure on banks, insurance companies, and critical infrastructure.

Furthermore, the concentration on a few technology providers increases exposure to common flaws, creating systemic effects capable of spreading rapidly.

Principle	Brief Description
Resilience	Maintaining stable services even during outages or attacks.
Governance	Define clear controls, responsibilities, and supervision.
Safety	Testing models against abuse, leaks, and security loopholes.
Transparency	Explain the limits, risks, and criteria for use.

Therefore, the FSB recommendation seeks to align innovation and prudence, ensuring that the use of AI is accompanied by robust testing, continuous monitoring, incident protocols, and regulatory adaptability.

Thus, technological advancement occurs without compromising global stability.

In summary , it is crucial that the global financial system addresses the challenges posed by cyber risk linked to AI.

The responsible and secure implementation of advanced technologies is fundamental to maintaining market confidence and ensuring the resilience of financial institutions.