Sysdig Japan LLC
Sysdig extends its AI workload security capabilities to AWS AI services Amazon Bedrock, Amazon Q, and Amazon SageMaker security improves AI visibility, identifies active risks, and helps AWS users meet AI compliance
……
-This press release is an abridged translation of the release announced in the United States on 6/10//2024-
6/10/2024 – PHILADELPHIA, PA (AWS re:Inforce) — Sysdig, the leader in cloud security powered by runtime insights, today announced that it will deploy AI workload security to Amazon Bedrock, Amazon SageMaker announces expansion to Amazon Q.
[Image 1: https://prtimes.jp/i/107539/18/resize/d107539-18-bece889a49b8c862e951-0.png&s3=107539-18-76135ac021ee3ed925a50ddb9cf615e3-3900×2041.png ]
In a world where it is difficult for security teams to stay ahead of attackers, AI workloads that contain large amounts of sensitive training data are being targeted. AI Workload Security, AI Workload
Security (https://sysdig.jp/press-releases/sysdig-launches-ai-workload-security/) is an extension of the Sysdig Cloud Native Application Protection Platform (CNAPP) that enables Identify and manage AI risks, giving security teams greater visibility into their environments and identifying suspicious AI workloads in real-time.
Amazon Bedrock, Amazon SageMaker, and Amazon Q support generative AI by providing users with high-performance foundational models (FMs) and the flexibility to innovate generative AI applications that are fully integrated into AWS environments. simplifies the development of application-based applications. As of last month, AWS announced that more than 10,000 organizations around the world are leveraging these AWS AI services. However, generative AI workloads come with security risks. Sysdig’s threat research team found that generative AI workloads are 35% more likely to be exposed than regular workloads. Elevated exposure risks and lack of visibility not only slow down the pace of software development, but also increase risk when particularly vulnerable applications are released into production.
Achieving faster, safer innovation
The cloud is faster, more complex, and more dynamic than on-premises environments, making it more susceptible to attacks. AI further complicates these security risks. Companies have only 5 seconds to detect an attack, 5 minutes to investigate, and 5 minutes to respond(*1). Sysdig and AWS are innovatively working to help customers deploy AI securely and quickly. AWS streamlines the process of building and scaling AI, and Sysdig’s unique capabilities of real-time detection and detailed runtime visibility help detect suspicious activity within these workloads and address the most pressing threats. Masu.
*1: Sysdig research: Reference:
https://sysdig.jp/wp-content/uploads/jp-sysdig-5-5-5-cdr-framework.pdf By extending AI Workload Security to AWS AI services and capturing real-time signals from AWS CloudTrail logs, Sysdig helps you reduce the impact and quickly respond to events such as:
Reconnaissance: Discover and detect attempts to exploit AI services, allowing security teams to prevent fraud before it occurs.
Data tampering: Identify attempts to manipulate data, delete models or knowledge bases, or disable logging to ensure the protection of sensitive data and the integrity of AI applications.
Internet exposure: Highlight where your AI applications are exposed to the internet, giving your team the visibility they need to limit the exposure of their own sensitive information.
Figure: Screen showing a risk found on Amazon BedRock and its attack vector
[Image 2: https://prtimes.jp/i/107539/18/resize/d107539-18-5aa4af030f52aa075a2a-1.png&s3=107539-18-a4b4a4b7ecc96da75e9df05324d7e750-3024×1890.png ]
Comment from Loris Degioanni, CTO and Founder of Sysdig Inc. “Everyone is eager to incorporate AI into their software, but incorporating AI without properly understanding the risks and implementing appropriate security measures can be costly. We are collaborating with AWS to enable our customers to securely take advantage of the efficiency and speed that AI brings.”
As the developer of Falco, the open source standard for cloud threat detection, Sysdig understands the importance of speed in attack response. By increasing visibility into applications that incorporate AI clients to communicate with AI services, Sysdig enables you to manage and control both legitimate and malicious AI usage. Sysdig streamlines triage and reduces response time by integrating real-time AI workload security with the company’s unified risk discovery capabilities. The solution provides security teams with a unified view of all relevant risks and events, enabling more efficient workflows for prioritizing, investigating, and mitigating active AI risks. notice:
Sysdig Japan will be a silver sponsor of AWS Summit Japan 2024 (https://aws.amazon.com/jp/summits/japan/), which will be held at Makuhari Messe from Thursday, June 20th to Friday, June 21st. (Booth ID: H4-S083) You can view a demo, including some of the AI workload security features announced today.
Reference materials:
Blog: AI Security in the Cloud: AI Workload Security on AWS: https://sysdig.jp/blog/ai-workload-security-for-aws/
Sysdig website “AI Workload Security”
https://sysdig.jp/ai-security-workflow/
Blog: Accelerating AI: AI Workload Security in CNAPP
https://sysdig.jp/blog/ai-workload-security-for-cnapp/
More details about this release:
https://prtimes.jp/main/html/rd/p/000000018.000107539.html