Home > News > Top Stories > Contrast Security Releases Assess Feature for LLMs to Protect Against AI Security Threats

Enterprise Security, Platform Security, Software Security

Contrast Security Releases Assess Feature for LLMs to Protect Against AI Security Threats

Contrast Security (Contrast), the code security platform built for developers and trusted by security, today announced it will extend its market-leading application security testing (AST) platform to support testing of Large Language Models (LLMs) from OpenAI. In this first release, Contrast rules help teams that are developing software using the OpenAI application programming interface (API) set to identify and mitigate weaknesses that could expose an organization to prompt injection vulnerabilities: i.e., attacks involving injection of a prompt that deceives the application into executing unauthorized code.

Prompt injection was identified as the top risk for LLM applications by the just-released OWASP 10 Top for Large Language Model Applications project. Contrast has continued to support OWASP's mission to improve Application Security (AppSec): In fact, Contrast's Chief Product Officer Steve Wilson led the 400-person volunteer team that created the OWASP Top 10 for LLMs.

"As project lead for the new OWASP Top 10 for LLMs, I can say our group looked deeply at many attack vectors against LLMs. Prompt Injection repeatedly rose to the top of the list in our expert group voting for the most important vulnerability," said Wilson. "Contrast is the first security solution to respond to this new industry standard list by delivering this capability. Organizations can now identify susceptible data flows to their LLMs, providing security with the visibility needed to identify risks and prevent unintended exposure."

According to the OWASP Top 10 for LLMs, a prompt injection vulnerability allows an attacker to craft inputs that can manipulate the operation of a trusted LLM. This results in the LLM acting as a "confused deputy" on behalf of the attacker. Given the high degree of trust usually associated with an LLM's output, the manipulated responses may go unnoticed and may even be trusted by the user, allowing the attack to potentially poison search results, deliver incorrect or malicious responses, produce malicious code, circumvent content filters, or to leak sensitive data. Prompt injections can be introduced via various avenues, including websites, emails, documents or any other data source that an LLM might rely on.

Contrast is ideal for identifying all types of injection accurately, including this new form of AI prompt injection. Contrast uses runtime security to monitor actual application behavior and detect vulnerabilities, rather than scanning source code or simulating attacks. This approach is fast, easy and highly accurate, ensuring that developers are instantly notified of issues and provided all the information they need to correct problems. User input sent through OpenAI's official Python API to an LLM in a Python agent-instrumented application triggers the prompt injection rule.

About Contrast Security

Contrast is a world-leading code security platform company purposely built for developers to get secure code moving swiftly and trusted by security teams to protect business applications. Developers, security and operations teams quickly secure code across the complete Software Development Life Cycle (SDLC) with Contrast to protect against today's targeted AppSec attacks.

Founded in 2014 by cybersecurity industry veterans, Contrast was established to replace legacy AppSec solutions that cannot protect modern enterprises. With today's pressures to develop business applications at increasingly rapid paces, the Contrast Secure Code Platform defends and protects against full classes of Common Vulnerabilities and Exposures (CVEs). This allows security teams to avoid spending time focusing on false positives, leaving them more time to remediate true vulnerabilities faster. Contrast's platform solutions for code assessment, testing, protection, serverless, supply chain, APIs and languages help enterprises achieve true DevSecOps transformation and compliance.

Contrast protects against major cybersecurity attacks for its customer base, which represents some of the largest brand-name companies in the world, including BMW, AXA, Zurich, NTT, Sompo Japan and the American Red Cross, as well as numerous other leading global Fortune 500 enterprises. Contrast partners with global organizations such as AWS, Microsoft, IBM, GuidePoint Security, Trace3, Deloitte and Carahsoft, to seamlessly integrate and achieve the highest level of security for customers.

The growing demand for the world's only platform for code security has landed the company on some of the most prestigious lists, including the Inc. 5000 List of America's Fastest-Growing Companies and the Deloitte Technology Fast 500 List of fastest-growing companies.

Spotlight

11 Ways to Streamline SEC Cybersecurity Compliance

Cyber attacks are increasing in volume, sophistication, and severity, and the federal government has taken notice. Now, they’re taking action — most recently in the form of new cybersecurity rules from the Securities and Exchange Commission. In “11 Ways to Streamline SEC Cybersecurity Compliance with Risk Cloud,” we explore a va

More featured news

Spotlight

11 Ways to Streamline SEC Cybersecurity Compliance

Related News

Software Security

Okta Unveils Identity Threat Protection with Okta AI for Its Cloud

Okta | October 09, 2023

Okta, Inc. announces Identity Threat Protection with Okta AI, a new product for Okta Workforce Identity Cloud. It enables security teams and admins to constantly evaluate user risk during active sessions and respond automatically to Identity threats across the entire ecosystem. Identity Threat Protection enables organizations to mitigate risk with enhanced threat detection and response capabilities. Okta, Inc., the leading independent Identity partner, has introduced Identity Threat Protection with Okta AI, a revolutionary product within the Okta Workforce Identity Cloud. This innovative solution provides identity-based threat detection and mitigation in real-time. Identity Threat Protection goes beyond initial authentication by employing Okta AI and security infrastructure insights. It assures continuous user risk assessment during active sessions and enables automatic responses to Identity threats across the entire ecosystem. Organizations are increasingly adopting diverse cybersecurity tools to counter evolving threats. However, this leads to a flood of detailed security data, making it daunting for administrators to set up efficient policies and identify crucial threats. This fragmentation makes it challenging to track the risk of any specific user session over time and requires navigating various consoles. Okta stands out with its unified Identity deployment, offering a unique ability to evaluate risks across security domains and active user sessions. This innovative strategy solves a crucial business challenge as, according to Gartner, "By 2025, organizations that adopt a continuous adaptive trust strategy will reduce [account takeover] along with other identity risks by 30%..." President and CDO of Workforce Identity Cloud at Okta, Sagnik Nandy, said, Organizations need the ability to not only bring together risk insights at the point of login, but also to re-evaluate at any point in a user’s session. Identity Threat Protection extends Okta’s adaptive risk analysis and delivers automatic remediation and response, assisting businesses stop potential threats in real-time. [Source – Business Wire] Multi-factor authentication (MFA) is a vital tool in combating identity-based attacks. However, its effectiveness typically extends only to the login stage. Using shared signals throughout a user's active session, Identity Threat Protection equips organizations with enhanced threat detection and response capabilities, thereby mitigating risk. Initial launch capabilities include: Continuous Risk Evaluation Shared Signals Pipeline Adaptive Actions CIO at Palo Alto Networks, Meerah Rajavel, said, Our world-class threat intelligence innovations assist secure enterprises against evolving attack methods. Partnering with Okta’s Identity Threat Protection is a great opportunity to further empower enterprises with security signal sharing, assisting them to better detect changes in user risk across their tech stack. [Source – Business Wire]

Enterprise Security

Microsoft harnesses power of AI to boost Windows 11 security, pushes for passwordless future

VentureBeat | September 27, 2023

Microsoft announced a series of new enterprise security features today that use artificial intelligence (AI) to help defend Windows 11 against increasingly sophisticated cyberattacks. The tech giant claims its new AI capabilities will reduce security incidents by 60% and firmware attacks by 300% for businesses using the latest version of its software. Microsoft’s vice president of enterprise and OS security, David Weston, explains in a company blog post that was published today specifically how AI is being used to fortify Windows 11 against sophisticated attacks, ranging from malware to firmware threats, and even nation-state attacks. At the heart of this AI-focused security upgrade is the integration of Microsoft’s Pluton Security Processor and Secured-core PCs. Both systems leverage AI algorithms to isolate sensitive data and provide defense against potential threats. IT professionals should note that these Secured-core PCs are reported to be 60% more resilient to malware than non-Secured-core PCs, a significant improvement in system defenses. Microsoft’s AI strategy also appears to be forward-thinking, with the company starting to adopt memory safe languages like Rust for traditional attack targets. Rust’s memory safety features without garbage collection make it an ideal language for building reliable and efficient systems, further multiplying the cybersecurity benefits. The end of passwords? Microsoft’s groundbreaking move Perhaps most notable among today’s announcements is Microsoft’s push towards a passwordless future, a move that could fundamentally alter the landscape of cybersecurity. Microsoft’s AI will now be used to develop and implement passkeys — unique, unguessable cryptographic credentials securely stored on a user’s device, which have the potential to replace traditional multi-factor authentication. This is a substantial leap forward in phishing protection, making it considerably more difficult for hackers to exploit stolen passwords. Microsoft says that its AI system analyzes more than 65 trillion security signals per day— with more than 4,000 password attacks every second — to identify suspicious login attempts and request additional identity verification when needed in the new system. The company also revealed a new capability called Config Refresh that relies on AI to detect and revert unwanted changes to device policies in near real-time. This allows IT teams to lock down device settings while leveraging intelligence to accommodate legitimate policy updates. Microsoft pioneers a new cybersecurity path with AI The company’s commitment to AI solutions aligns with its longstanding strategy of positioning itself as a leader in enterprise computing. By weaving AI into the fabric of Windows 11, Microsoft is demonstrating its commitment to providing businesses with secure, reliable, and forward-thinking solutions. Business analysts see this as a clear indicator of Microsoft’s strategy to leverage its AI prowess to drive growth and cement its position in the enterprise data and AI market. Given the increasing importance of cybersecurity in the modern business landscape, Microsoft’s investment in AI could pay significant dividends. However, the real test of these new AI-powered features will be their effectiveness against real-world threats. As cyber threats continue to evolve, so too must our defenses. If Microsoft’s AI enhancements can live up to their promise, they will represent a significant advance in cybersecurity and a potent tool in the fight against cybercrime. As AI continues to transform enterprise data and security, it’s clear that companies like Microsoft are leading the charge. By harnessing the power of AI, Microsoft is not just shaping its future but also the future of cybersecurity as a whole. Only time will tell how these developments play out, but one thing is certain: the era of AI-driven cybersecurity is here, and Microsoft is at its helm.

Data Security

Oracle Attempts to Design New Open Network and Data Security Standard

Oracle | September 20, 2023

Oracle to participate in an industry-wide initiative to design a new open network and data security standard. Oracle and Applied Invention are assisting to developing and promoting a novel network and data-centric security standard to tackle distributed cloud deployment challenges. This standard will enable organizations to protect their data throughout its entire lifecycle without requiring modifications to their distributed cloud environments' underlying architecture. Oracle, one of the world's largest database management companies, announced that it will participate in an industry-wide initiative to design a new open network and data security standards that will assist organizations in protecting their data in distributed IT environments. Oracle will collaborate with Applied Invention, a significant technology provider, and other industry leaders, including Nomura Research Institute, Ltd. (NRI), a global leader in consulting and system solutions. This new standard will enable networks to enforce shared security policies collectively, thereby augmenting the security architecture organizations already employ without requiring modifications to existing applications and networks. Oracle plans to launch the Oracle Zero-Trust Packet Routing Platform, based on the new standard, to support this new initiative. This platform will assist organizations in preventing illegal access or use of their data without imposing additional obstacles on legitimate activities. Executive Vice President of Security and Developer Platforms at Oracle Cloud Infrastructure, Mahesh Thiagarajan, said, Over the last 20 years, the cybersecurity industry has produced many incremental changes, but we need a fundamentally novel approach to protect our data in the increasingly complex cloud era. Organizations require a way to describe their data security policies in one place where they can be easily understood and audited, and they need a way to make sure those policies are enforced across their entire computing infrastructure, including their clouds. [Source – Cision PR Newswire] As the adoption of cloud technology rises and IT landscapes become more intricate with distributed cloud deployments, organizations face escalating challenges in safeguarding their data using conventional methods and tools. For example, many existing systems necessitate security teams to orchestrate disparate solutions across various facets, including database, application, network, and identity security. This complexity is further compounded when applied across diverse environments. Ensuring seamless collaboration among these solutions becomes a formidable task due to the dynamic and independent changes in applications, environments, and user profiles. Additionally, current security systems demand extensive configurations to accurately distinguish between different user categories, such as full-time employees and contractors, without compromising security or restricting access. Research Vice President of Cloud and Edge Infrastructure Services at IDC, Dave McCarthy, said, The new standard Oracle develop has the potential to change all of that by adding a unified layer of security on top of existing solutions. Building data protection policies into the network itself will assist users get the access they require while ensuring the data remains secure behind the scenes. [Source – Cision PR Newswire] Oracle and Applied Invention are assisting in designing and promoting a novel security standard, focusing on network and data-centric security, which aims to tackle these challenges. This innovative standard will empower organizations to safeguard their data across its entire lifecycle, including distributed cloud environments. To accomplish this, the standard will implement an intent-based security policy that is designed to be understandable, auditable, and interpretable by humans. This intent-driven approach will be put into practice at the network layer, ensuring that every data transmission contains authenticated attributes concerning the sender, receiver, and the nature of the data in transit.