The Fight to Keep our Data Safe During the Age of AI
Data is the future and protecting it in the current AI landscape is essential.
With business demand for new technologies booming like never before, data has also experienced rapid growth. By 2025, McKinsey predicts that nearly all employees will regularly leverage data to support their work. Every day, the world produces five exabytes of data, and by 2025 this is set to rise to a rate of 463 exabytes, driven by increased adoption in AI.
However, as businesses continue to embrace AI more frequently, they are also increasingly concerned about how the technology could affect their valuable data.
AI sparks huge data growth
Surging growth of data in 2024 is central to the current global AI landscape. Companies in the technology sector like Dell Technologies and AMD for example are working to ensure that organisations have access to the compute power needed to fuel their innovations. As a result, they can offer customers more choice to power AI and generative AI (Gen AI) workloads.
The rapid evolution of AI ultimately presents huge opportunities for companies and their data. However, they must also prepare to deal with increased customer demand for data in line with data regulations.
“AI is unique, often presenting challenges, particularly in the context of data privacy,” comments Ramprakash Ramamoorthy, Director of AI Research at ManageEngine: the enterprise IT management division of Zoho. “AI’s ability to process and analyse large volumes of data can sometimes run counter to the principles of data privacy, creating a scenario where AI is both the facilitator and the challenger in this domain.
“The ongoing challenge lies in developing AI systems that are not only efficient in data processing but also inherently respect and protect privacy.”
In fact, a January 2024 study conducted by Informatica has found that while a new era of Gen AI promises more potential for data leaders, it also brings with it increased challenges for businesses.
The study confirmed that more than two in five (43%) Gen AI adopters surveyed cited AI governance as the main obstacle, in addition to AI ethics and the quality of domain-specific data for training large language models (LLMs).
In order to ensure a successful AI strategy in the year ahead, many businesses are planning to prioritise investments into data management capabilities so that they are better prepared for AI. Additionally, moving towards AI-powered data management has been suggested to better help businesses best demonstrate compliance with regulations and lay the foundations for wider generative AI adoption.
Jason du Preez, Vice President of Privacy and Security at Informatica, comments on why the companies that collect our data have the power and the responsibility to use it for good. “Consumer expectations are moving compliance and the responsible use of data higher up the agenda - both with today’s privacy regulations and future, AI-focused legislation,” he says. “Data governance, security and privacy are critical business priorities as a result.
“Responding to this, and then going further to drive innovation and competitive advantage with data, requires a holistic approach to data management. Understanding, protecting, distributing, and monitoring data universally and consistently and ubiquitously becomes a necessity.”
Protecting digital assets from cybercrime
When considering how quickly data has grown, it is important to see why businesses must ensure that their data protection strategies are up-to-date to protect against so-called threat actors.
In recent years, AI is being increasingly used to extort data from individuals and businesses of all sizes. It is expected that cybersecurity threats will become more personalised as AI uses continue to evolve, providing more opportunities for threat actors to obtain valuable information.
In particular, the rise of deep fakes poses a significant threat to identity and access management, requiring organisations to prioritise solutions that verify genuine human presence and implement additional security measures.
A February 2024 report by Gartner predicts that by 2026, attacks using AI-generated deepfakes on face biometrics will mean that 30% of enterprises will no longer consider such identity verification and authentication solutions to be reliable in isolation.
Cyberattacks are evolving and unfortunately inevitable. As a result, industry leaders and businesses will need to consider new and innovative ways to counter cyber threats to data.
“Looking ahead, AI will continue to face challenges from its own advancements,” Ramamoorthy says. “With the rise of more sophisticated AI technologies, such as generative models, managing privacy concerns only becomes more complex. The future of AI will involve creating self-regulating algorithms that can identify and mitigate privacy risks autonomously.
“The evolution of AI will likely see the development of more advanced privacy-preserving techniques like differential privacy and homomorphic encryption.”
The future of ‘safe AI’
However, despite these concerns, AI can be used for good if harnessed correctly. If businesses have stringent data privacy practices and ethical safeguards in place, then they will be more prepared to counteract data breaches in the long run.
On this, Shankar Haridas, UK Head of ManageEngine, says: “Business leaders and stakeholders often underestimate the value of the data they maintain until they face a breach. It only takes one loophole for an attacker to access data, and unless an organisation is proactively reviewing its security posture, it's at real risk of attack, with a response that's often taken too late.
Haridas advises that organisations should be investing in data loss prevention solutions that are able to monitor and prevent sensitive business data from leaking. “Beyond implementing stronger security measures, training staff on best practices and identifying clear red flags helps protect data,” he explains
“With AI being integrated into every aspect of business, disruptive technologies, like deep fakes and augmented reality, threaten data privacy and pose a significant risk to business,” Haridas concludes. “It’s increasingly important that regulators and businesses strike a balance that leverages the benefits of AI while limiting its destructive capacity.
“Data privacy is essential when it comes to policies around AI.”
******
Make sure you check out the latest edition of AI Magazine and also sign up to our global conference series - Tech & AI LIVE 2024
******
AI Magazine is a BizClik brand
