Data is the oil of our times. This idea is not merely a snappy tagline—it is a fundamental truth. Data is stored, processed, and analyzed in new and innovative ways thanks to advancements in cloud computing, artificial intelligence (AI), machine learning (ML), and the Internet of Things (IoT). While these advancements offer many benefits, they also create risk. Securing data is a highly challenging task and is quickly becoming a top priority for organizations as emerging technologies continue to advance and shape the way data is handled. Organizations must be aware of several areas of risk to secure valuable data, and there are several solutions that organizations can implement to protect sensitive information and systems.
Data Security Challenges Due to Emerging Technologies
With the increasing integration of new technologies, organizations must implement robust data security measures to protect information. Technologies such as IoT and AI help organizations work faster and smarter, but also create new vulnerabilities for hackers to exploit. For instance, in 2021, Colonial Pipeline suffered a ransomware attack that caused major issues with fuel supply in the United States.1 This incident demonstrates that new technologies can introduce new security problems. This means that organizations need to update their protection methods to keep pace with these changes.
Lack of Cloud-Based Data Security
It is important to distinguish between the security of the cloud and security in the cloud. The security of the cloud is the responsibility of cloud service providers who manage infrastructure and ensure it is physically and logically secure. However, security in the cloud depends on the user and their actions. This means that organizations must actively implement cybersecurity measures such as secure configuration, data encryption, access controls, and regular vulnerability scanning to secure both systems and sensitive consumer data.
Moreover, when organizations commit to adopting cloud-based solutions, few move their entire IT infrastructure and data systems at once. Instead, they prioritize migrating certain systems, datasets, or use cases, allowing them to efficiently manage costs, minimize operational disruptions, and test the security and performance of cloud environments before fully committing to any one solution. Despite the benefits of a phased approach, the adoption of cloud-based data security capabilities, such as data loss prevention, is often deemed low priority, leading to a lag in security practices. Moreover, multicloud environments pose the risk of unencrypted data transfer and data leakage. These security gaps increase the risk of misconfigurations and data leaks, making it essential for organizations to enforce merged cloud security policies in hybrid environments.
AI-Generated Malware and Ransomware Attacks
AI's ability to automate malware development enables attackers to identify and exploit system vulnerabilities faster than ever before. This emerging capability could lead to a rise in sophisticated cyberattacks that exploit zero-day vulnerabilities, posing grave threats to organizational security.
Similarly, ransomware continues to be one of the most disastrous threats to organizations.2 In these attacks, malicious actors gain access to critical organizational data and demand a ransom to reinstate an employee's access. Cloud-hosted software as a service (SaaS) applications and storage systems are a valuable target, as they are frequently used and thus present serious security vulnerabilities.
IoT-Related Risk
IoT offers great potential but comes with novel risk. This is why modern security measures are quickly replacing traditional encryption methods. IoT devices are a major target of cybercriminals, and many of these devices are inadequately secured and easy to breach. Attackers mostly exploit default passwords, outdated firmware, and unpatched vulnerabilities to gain access. Once they have successfully infiltrated a system, attackers can hijack the device, use it as a launch point for larger attacks, or extract sensitive data from connected networks.
Advanced Persistent Threats (APTs)
The workloads distributed across private, public, community, and hybrid infrastructure have enabled hackers to access systems through multiple potentially vulnerable entry points. APTs refer to prolonged and targeted cyberattacks carried out by skilled threat actors who often infiltrate networks silently and remain undetected for extended periods. These threat actors exploit data through prolonged attacks and human error to infiltrate networks, software, and weak access controls.
Generative AI-Based Attacks
Organizations' data security policies are falling behind the fast pace of AI development, and enterprise applications are increasingly at risk. Today, many employees seek assistance from unauthorized generative AI tools to enhance productivity and automate daily tasks.3 However, while using these tools, employees may unknowingly share sensitive or proprietary data with the generative AI models, which can then be exposed through model leaks or exploited by threat actors. Another AI-based attack, data poisoning, is a dataset manipulation attack, where adversarial triggers change the input data. This can degrade model performance, insert hidden backdoors, or bias outputs, helping attackers disrupt operations or influence security decisions. Data exfiltration is another challenge for AI systems, as it affects data security. Attackers may use generative AI systems to extract confidential data, posing a serious threat to enterprise data security.
Addressing Data Security Risk in Emerging Technologies
Though technology is evolving at a rapid pace, there are several strategies organizations can implement to ensure the safety and integrity of sensitive data:
- Implementing zero trust architecture (ZTA)—ZTA is based on not implicitly trusting any user, device, or application and consistently validating their interactions at every layer. Zero trust policies include enterprisewide microsegmentation, identity governance, role-based access, and the strict enforcement of the principle of least privilege to ensure just-in-time (JIT) and just enough access (JEA) to data.
- Deploying AI-powered monitoring tools—Real-time monitoring tools analyze large volumes of data to detect suspicious patterns, such as unusual logins and data transfers, to take swift action and nullify the chances of a breach. AI can automate vulnerability testing and patching, freeing employee time for more strategic work. Organizations must constantly refresh their AI models to prepare for AI-powered attacks. Real-time AI-driven monitoring systems can analyze large volumes of network and user activity data to detect anomalies that may indicate a security threat, such as unusual login attempts, large unauthorized data transfers, or access from suspicious IP addresses. Vulnerabilities and risk associated with this strategy include model drift and obsolescence, false positives or negatives, and adversarial attacks.
- Enable auditing for all sensitive fields—Audit logs capture detailed insights, including field updates, changes to values, timestamps, and user login activity. Enabling auditing for all sensitive fields helps security personnel detect unauthorized activity and identify data discrepancies more effectively.
- Enforce DLP policies—Create data loss prevention (DLP) rulesets to protect sensitive data from unauthorized access and inadvertent sharing with external systems that are outside of the organization's firewall or network boundaries. These rules help locate, track, and automatically limit the migration of sensitive data across email, cloud storage, and other channels.
- IoT security measures—Mechanisms such as IoT device authentication, edge network segmentation, logging, periodic firmware updates, and AI-driven analytics should be deployed when monitoring IoT traffic for compromise signals. This measure is particularly important due to the increasing adoption of IoT devices. Moreover, organizations should consider a centralized dashboard for IoT management that allows IT administrators to monitor, configure, and update all connected IoT devices from a single interface. This proposed dashboard can provide better visibility and oversight into all connected devices, enabling real-time tracking of device status, security events, and traffic patterns.
- Bolster employee training and organizational culture—Human error plays a major role in data breaches. Research shows that a significant percentage of cybersecurity incidents are caused by employee mistakes.4 To avoid this common organizational pitfall, employee training should be interactive and include ransomware and phishing simulations, gamification, and security awareness training on emerging threats.5 Organizations should build a security-first culture in which all employees are committed to protecting sensitive data.
- Strong incident response and recovery plans—An incident response plan reduces the impact of a data breach and speeds up recovery. The incident response team should be ready with well-developed workflows for novel scenarios that could result from emerging tools. The team should also facilitate regular drills and update response protocols to prepare for emerging threats. Moreover, enterprises should invest in secure backup solutions and test a rollback procedure for the quickest data restoration in case of an attack. This rollback procedure ensures that data can be swiftly and accurately restored in the aftermath of a cyberattack or data breach. It involves simulating scenarios in which critical data is compromised, deleted, or encrypted, such as in the case of ransomware.
Conclusion
With organizations' attack surfaces exponentially expanding, there is an increased need for a security-first, innovative organizational culture. To accomplish this, organizations must focus on continuous improvement and bolstering resilience while implementing strategies that shift security to the left. Dynamic strategies for embedding security in emerging technologies, such as AI, IoT, and cloud computing, will require industrywide collaboration alongside developments in the regulatory landscape to set up robust security frameworks.
The future belongs to those who raise the bar regarding data confidentiality, integrity, and scalability standards, as well as those who better position themselves to thrive in a world where security challenges are inevitable.
Endnotes
1 US Cybersecurity and Infrastructure Security Agency (CISA), “The Attack on Colonial Pipeline: What We’ve Learned & What We’ve Done Over the Past Two Years,” USA, 2 May 2023
2 Statista, “Ransomware – Statistics and Facts,” 19 November 2024
3 Downie, A.; Hayes, M.; “AI in the Workplace: Digital Labor and the Future of Work,” IBM, 16 October 2024
4 Peters, J.; “Human Error is Responsible for 74% of Data Breaches,” Infosec, 30 November 2023
5 Bitrián, P.; Buil, I.; et al.; “Gamification in Workforce Training: Improving Employees’ Self-Efficacy and Information Security and Data Protection Behaviours,” Journal of Business Research, vol. 179, iss. 114685, 2024
Aparna Achanta
Is a seasoned security architect and leader at IBM Consulting with extensive experience driving mission-critical cybersecurity initiatives, particularly in federal agencies. She successfully implemented cybersecurity frameworks, such as zero trust and security by design for federal clients, strengthening the security posture and enhancing data protection and security standards across cloud applications. Achanta specializes in securing emerging technologies in federal agencies, including low-code and no-code applications and generative AI applications.