Bosch AIShield was recently unveiled at CES 2022 as an industry-first & patented AI Security SaaS offering to protect AI systems against emerging security threats
By Manoj Parmar, Global Program Director – AIShield at Bosch, Amit Phadke, Global Product Manager – AIShield at Bosch
Every organization wants to protect its valuable digital assets like data and information, but what about AI algorithms? Successful exploitation of AI algorithms can cause financial loss, reputational damage, loss of competitive advantage, and loss of intellectual property. For example, a malicious breach of pricing algorithm for an e-commerce company can result in attackers causing financial loss to the retailers/sellers and negatively impacting the platform’s brand reputation. Similarly, a heavily invested medical diagnosis AI algorithm, when stolen or attacked, could lead to loss of niche competitive advantage and loss of patient data. Similar challenges & threats have also been highlighted recently by enisa (EU agency for cybersecurity) and NIST (National Institute of Standards & Technology) with Bosch being a key comment contributor to the later. Bosch has been researching & working on this as well since 2020.
According to Bosch, the majority of organizations are struggling to address the challenge of securing models. Suppose one wants to protect their algorithm. In that case, they must check the latest research related to an adversarial aspect of AI, find the right people with the proper knowledge of AI and security, figure out how to use open-source research in the development of algorithms, and then understand how to integrate with the development workflow easily and intuitively. Lack of addressing these challenges results in unanswered questions such as – how often your algorithm is really under attack? What is the true financial value of AI models, and what if they are stolen? What’s the impact of the AI model attack on brand reputation and customer trust?
Therefore, the need of an hour is to create a last layer of defense for AI models to plug in the newer attack surfaces.
Bosch AIShield brings hacker-level vulnerability assessment and security hardening to the organization’s AI-powered devices and cloud solutions. The product provides a SaaS-based
a tool with UI so that every AI stakeholder (e.g., data scientist, ML engineers, product managers, security engineers) can benefit from our patient-powered deep technology. With a few clicks, AI stakeholders can get vulnerability assessment reports, generate, and deploy customized defense mechanisms, and get real-time notifications on real-time threat detection.
“Bosch is committed to making AI trustworthy and build digital trust in AI. AIShield is an important product offering towards that. It protects AI systems against emerging threats from adversaries. AIShield makes AI security accessible to all and bolsters Digital trust in AI”, says Krishnamurthy Vadiraj, Head of Technology & Innovation, BGSW.
Bosch AIShield also offers out-of-the-box native automation support so that organizations can achieve scale rapidly. With a minimum configuration, it can be easily, effortlessly, and hassle-free way integrated into the MLOps (ML workflows) leveraging our API offerings. Bosch AIShield takes an AI model and sample data as input and calculates the vulnerability score of a model using our intelligent attack framework and attack database. The attack database is foundational and constantly updated through our research and open-source intelligence. Bosch AIShield identifies relevant attack vectors and, using them, calculates the vulnerability score. This score prioritizes the most critical vulnerabilities and creates a defense mechanism against attacks as a remediation response. Bosch AIShield generates a deployable version of the defense model and real-time attack notifications. The notifications are designed to be easily be integrated by industry-leading Security Incident and Event Management services. The user interface provides intuitive visualizations and detailed reports. Bosch AIShield also offers consulting services to help the organization navigate the security landscape of AI. With Bosch AIShield’s rich end-to-end security offering, even small security teams and AI/ML developers can ensure AI systems are monitored and protected.
About the Authors
Amit is a seasoned software professional and an experienced product manager with varied experience of 9+ years in building products for the international automotive market. He currently is the Global Product Manager for AIShield and has experience managing diverse distributed teams. One of his past roles as Executive Assistant to Senior Vice President of Innovation has given him keen insights into the innovation journey of an empowered captive center of a large MNC tasked with building products for the globe. He has completed an executive education certificate program from UCLA, Anderson School of Management (PGP Pro). His areas of interest are Product Management, Innovation, Strategy, and Agile Development.
Manoj’s role is Global Program Director – AIShield® at Bosch. He is an award-winning, experienced, seasoned Technologist. He has been working at Bosch for more than 13 years. His responsibilities include building a global product at the intersection of AI and Cybersecurity to safeguard AI systems and leading the AI Security Initiative for Bosch Group. Manoj’s journey is inspirational and transformational across automotive, two-wheeler, and digital domains to deliver customer-focused business values. Along with his teams, he has built several innovative products and solutions using multiple classical and emerging technologies.
He has filed 20+ patents and has published 13+ research papers. He is a mentor for deep tech startup events. He has also been a guest speaker for technology, innovation, and entrepreneurship.
He is a recipient of the Chevening Cybersecurity Fellowship 2021 by the UK Government’s FCD Office. Zinnov awarded him Technology Role Model Award 2020. He is a founding member of MITRE ATLAS. He holds engineering and management degrees. He has also completed the Corporate Startup program from UC Berkeley and M.Sc. in Innovation and Entrepreneurship from HEC Paris.
He is a lifelong learner and enjoys cooking, reading, and jogging.
To learn and engage more on Bosch AIShield, please visit:
AIShield Product listing on Azure Marketplace
AIShield with Azure Sentinel on Marketplace
FAIR USE NOTICE: Under the “fair use” act, another author may make limited use of the original author’s work without asking permission. Pursuant to 17 U.S. Code § 107, certain uses of copyrighted material “for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.” As a matter of policy, fair use is based on the belief that the public is entitled to freely use portions of copyrighted materials for purposes of commentary and criticism. The fair use privilege is perhaps the most significant limitation on a copyright owner’s exclusive rights. Cyber Defense Media Group is a news reporting company, reporting cyber news, events, information and much more at no charge at our website Cyber Defense Magazine. All images and reporting are done exclusively under the Fair Use of the US copyright act.
Source: www.cyberdefensemagazine.com