During a recent meeting with President Joe Biden, leaders from major technology giants like Amazon, Google, Meta, and Microsoft made a commitment to adopting safeguards for artificial intelligence (AI) systems to ensure their safety before releasing them to the public. One specific area of concern is cybersecurity, as the use of generative AI has made ransomware attacks and phishing schemes easier to execute and more widespread. Chief Information Security Officers (CISOs) and other cyber leaders are well aware of this growing risk and the democratization of cyber vulnerabilities. Collin R. Walke, head of cybersecurity and data privacy practice at law firm Hall Estill, noted that the rapid adoption of generative AI has tremendously changed the threat landscape, allowing more individuals with hacking capabilities to exploit cybersecurity weaknesses using tools like ransomware-as-a-service and AI. For instance, generative AI has made phishing attacks more convincing and authentic. Previously, employees could detect fake emails as they often contained wording that seemed off. However, with generative AI, non-English-speaking malicious actors can instantly and almost flawlessly translate phishing emails into any language, making it harder for employees to identify them as fake. However, experts point out that the same AI tools, which enable hackers to operate more efficiently and at a larger scale, are also accessible to companies seeking to strengthen their cybersecurity defenses. Stephen Boyer, co-founder and CTO of cyber risk management firm BitSight, emphasizes that while AI enhances attackers' capabilities, it also empowers defenders.

AI allows engineers to automatically check for vulnerabilities in code, resulting in more secure software. Boyer predicts that AI will significantly accelerate the process, compared to current tools available. The use of AI to increase speed and scale in cybersecurity is viewed as one of the notable advantages in the near future. Michael McNerney, Chief Security Officer at cyber insurance company Resilience, highlights the potential benefits of leveraging AI to streamline complex and time-consuming tasks faced by CISOs, such as device inventory and application management. However, caution is advised by Walke and others. Despite technology giants pledging to follow AI safeguards, there are concerns about AI companies worldwide continuing to develop and deploy the technology without adequate legal or ethical regulations. Therefore, cyber experts advocate for a collaborative approach, where CISOs work closely with the board, chief risk officer, and CEO to determine when and how to deploy AI across the organization. They emphasize the need for an "all-hands-on-deck scenario" to address the complexities and risks associated with AI, as witnessed by the recent data breach at OpenAI. It is important to maintain perspective and context when considering AI and its potential impact. McNerney suggests that while AI's hype is currently at its peak, it is natural for an emerging technology that is both exciting and powerful. Over the next year, cyber leaders will gain a better understanding of where AI can truly be advantageous and where it may not be as effective.