U.S. Treasury warns financial sector, AI threats “outpacing” security – Source: www.cybertalk.org

EXECUTIVE SUMMARY:

According to a new U.S. Treasury Department report, the financial services industry is extremely vulnerable to cyber threats that weaponize AI-based tools. The report provides warning to the industry at-large, while also sharing best practices and advocating for AI-based threat prevention.

The “…report builds on our successful public-private partnership for secure cloud adoption and lays out a clear vision for how financial institutions can safely map out their business lines and disrupt rapidly evolving, AI-driven fraud,” stated Under Secretary for Domestic Finance, Nellie Liang.

AI-powered attacks

According to high-level stakeholders who hail from financial technology companies, generative AI capabilities may give the “upper hand” to cyber criminals.

Experts anticipate that AI will supercharge malware potency, social engineering tactics, vulnerability discovery (on the part of hackers) and disinformation campaigns – including deepfake videos that show impersonation of executives.

Financial institutions have long utilized AI for cyber security, anti-fraud and other operational purposes. However, many have stated that their current risk frameworks remain inadequate when it comes to preventing novel artificial intelligence-based attack vectors.

As AI models become more resource-intensive, over-reliance on third-party cloud providers also presents new cyber security risks.

Short-term recommendations

The Treasury report details several immediate measures that financial services companies can take to mitigate risks:

• Leverage existing regulations. Although AI-specific rules are still emerging, many current cyber security, privacy and risk management regulations can be applied to AI system governance.
  
  
  
• Improve anti-fraud data sharing. At present, large banks have a major advantage in building AI fraud detection models, as they have large data reserves. More public-private data pooling is needed.
• Develop AI data supply chain mapping. Like nutrition labels for food, “AI nutrition labels” should be mandated to clarify the origins and parameters of training data used to build AI models.
  
  
  
• Cyber workforce transformation. Static training programs must be overhauled in order to develop AI-fluent cyber security professionals; talent that can effectively operationalize AI-based tools while upholding ethics, security and privacy standards.
  
  
  
• Push for increased government coordination. An inconsistent patchwork of state/federal AI rules presents a tangle of different challenges. Aligned regulations and public-private partnerships are needed in order to effectively combat threats.

Long-term solutions

In order to address the AI-based cyber security challenges outlined by the U.S. Department of the Treasury, financial institutions are also encouraged to explore Check Point’s industry-leading AI cyber security offerings.

Check Point’s unified cloud security platform secures financial AI workloads and data across public clouds, private clouds and on-premises, using comprehensive AI-powered security services.

Given how AI-based cyber threats are intensifying, banks and fintech groups need to urgently prioritize AI risk management programs, upgrading defenses before disruptive attacks manifest.

Early mover advantage

When it comes to getting ahead of the AI security curve, there is such thing as an early-mover advantage. By partnering with Check Point, financial institutions can acquire the strategic vision and execution velocity required to outpace modern threats. To learn more about AI-powered, cloud-delivered cyber security solutions, please click here.

In addition, empower your organization through more great thought leadership. Discover new artificial intelligence focused thought leadership insights from CyberTalk.org, here. Lastly, to receive cutting-edge AI cyber insights, groundbreaking research and emerging threat analyses each week, subscribe to the CyberTalk.org newsletter.