ChatGPT is reshaping shadow IT, introducing unforeseen risks for organizations. IT and cybersecurity leaders must leverage its speed while ensuring robust security. OpenAI reports that enterprise adoption is on the rise, with over 80% of Fortune 500 employees using ChatGPT.
A recent Harvard University study reveals that enterprise workers experience a 40% performance boost with ChatGPT. Additionally, MIT found that it reduces skill disparities and accelerates document creation, allowing workers to use their time more efficiently. However, 70% of employees are hesitant to disclose their use of the tool to management.
Are You Ready for AI Agents?
Reducing the risk of intellectual property loss while maintaining speed
The primary concern with ChatGPT is the potential for employees to inadvertently share sensitive information, such as intellectual property (IP), confidential pricing, financial data, and HR information. High-profile cases, like Samsung’s data leak, highlight the urgent need for effective solutions.
As organizations seek to address this issue, many are turning to generative AI-based approaches for enhanced security. Interest is growing in generative AI isolation technologies designed to keep confidential data away from ChatGPT, Bard, and similar platforms. Businesses aim to balance the efficiency and process improvements offered by ChatGPT with a robust risk management strategy.
During a discussion with Alex Philips, CIO of National Oilwell Varco (NOV), he emphasized the importance of educating the board on the opportunities and risks associated with generative AI. Philips regularly updates the board about the latest developments in generative AI and its application within NOV, helping to establish expectations and safeguards to prevent data leaks similar to Samsung's.
Innovative Solutions to Secure ChatGPT Sessions
A range of new technologies is emerging to address the challenge of securing ChatGPT sessions without compromising speed. Notable providers include Cisco, Ericom Security by Cradlepoint, Menlo Security, Nightfall AI, Wiz, and Zscaler.
These providers adopt various strategies to protect confidential data from being shared in ChatGPT sessions. Ericom Security by Cradlepoint’s Generative AI Isolation and Nightfall for ChatGPT are among the most widely recognized solutions.
Ericom employs a clientless approach, executing user interactions with generative AI sites in a virtual browser within the Ericom Cloud Platform. This method allows for data loss protection and access policy enforcement, ensuring sensitive information is not submitted to generative AI platforms like ChatGPT. Their design emphasizes delivering least privileged access via cloud architecture.
Nightfall AI offers a suite of three solutions aimed at safeguarding confidential data from exposure on ChatGPT and similar platforms. Nightfall for ChatGPT is a browser-based solution that scans and redacts sensitive information in real time. Nightfall for LLMs provides an API for detecting and redacting sensitive data used in training language models. Lastly, Nightfall for SaaS integrates with popular cloud services to prevent sensitive data exposure.
Generative AI: The Future of Knowledge
Generative AI is becoming the essential knowledge engine for businesses. Evidence suggests that banning ChatGPT and similar technologies counterintuitively fuels the rise of shadow AI, complicating data security efforts.
More CIOs and CISOs are now piloting generative AI systems, implementing strategies that mitigate risks at the browser level. By leveraging secure cloud architectures, as demonstrated by Cradlepoint Ericom, enterprises can protect thousands of employees from unintentional data sharing.
The focus should be on transforming the rapid advancement of generative AI into a competitive advantage. IT and security teams must prioritize understanding the latest technologies and methods for protecting sensitive information. Staying informed about these evolving protections is vital for maintaining competitiveness in the knowledge economy.
“Generative AI websites offer significant productivity gains, but organizations must be proactive in managing associated risks," said Gerry Grealish, Vice President of Marketing at Ericom Cybersecurity Unit of Cradlepoint. “Our Generative AI Isolation solution empowers businesses to balance the potential of generative AI with robust safeguards against data loss, malware threats, and legal compliance challenges.”
Ranking
