Cisco is happy to share that Sturdy Intelligence, a not too long ago acquired AI safety startup, has been talked about within the first ever 2024 Gartner Cool Distributors for AI Safety report.
As said by Gartner, “Safety and danger administration leaders face operational challenges, compliance considerations and assaults aimed toward AI techniques. This Cool Vendor analysis identifies suppliers with progressive methods of securing AI purposes, supporting AI belief, danger and safety administration capabilities.”
Furthermore, the inclusion of Sturdy Intelligence celebrates the progressive work the workforce finished since day one — creating the first-ever AI Firewall, present process in depth analysis into matters like superb tuning and algorithmic jailbreaking, and creating an end-to-end platform for AI safety which has change into the template for our business.
“We’re extremely excited and thrilled to be acknowledged within the Gartner Cool Distributors report,” mentioned Yaron Singer, co-founder of Sturdy Intelligence and VP of AI and Engineering at Cisco. “We predict this can be a testomony not solely to the groundbreaking work we’ve finished at Sturdy Intelligence, however extra broadly to the state of AI safety in the present day. In our view, when a number one international analyst agency like Gartner publishes a report like this, it reaffirms that AI safety is likely one of the most urgent challenges enterprises face in the present day.”
Why Sturdy Intelligence is a Cool Vendor for AI Safety
Unmanaged AI danger can introduce any variety of monetary, authorized, and reputational dangers for any enterprise adopting this transformative new expertise. The accountability of AI safety is shared by these creating AI purposes and the safety and governance groups defending delicate knowledge at an organizational degree. The best AI safety answer ought to facilitate safe and accountable growth whereas additionally imposing runtime controls after deployment.
As a pioneer on this house, Sturdy Intelligence launched the first-ever AI Firewall to the market as a part of their complete AI safety platform. Organizations on the reducing fringe of AI innovation belief Sturdy Intelligence to guard their AI purposes in opposition to a whole bunch of security and safety dangers.
The Sturdy Intelligence platform includes three core elements, defending AI purposes throughout their whole lifecycles:
- Mannequin file scanning proactively identifies safety vulnerabilities in open-source elements of the AI provide chain, comparable to fashions from Hugging Face. If undetected, these malicious insertions might in any other case compromise a whole AI system.
- AI Validation automates security and safety testing for AI fashions, serving to builders determine and mitigate these dangers earlier within the course of. This algorithmic pink teaming analyzes susceptibility to a whole bunch of assault strategies and menace classes for AI.
- AI Safety gives runtime safety for manufacturing AI purposes in opposition to all kinds of assaults and undesirable responses. This protection spans a whole bunch of classes and proceed to develop primarily based on cutting-edge AI menace analysis.
The engine behind the Sturdy Intelligence platform is powered by three proprietary parts: algorithmic pink teaming expertise, a world-class AI menace intelligence pipeline, and coverage mappings. The Sturdy Intelligence workforce has additionally helped set up AI safety requirements by means of our collaboration with MITRE, NIST, OWASP, and different entities. This deep expertise helps our prospects to mitigate all kinds of AI threats and monitor compliance with related insurance policies. It additionally permits us to ship on-the-fly updates to maintain our prospects protected against a dynamic menace panorama.
Sturdy Intelligence continues to be on the forefront of AI safety innovation, from creating the business’s first AI Firewall to conducting breakthrough AI analysis. Matters have included algorithmic jailbreaking, the dangers of fine-tuning basis fashions, the extraction of coaching knowledge, and exploits in a number of state-of-the-art LLMs.
Sturdy Intelligence’s expertise, built-in into the Cisco Safety Cloud, strengthens our skill to guard AI purposes and fashions in opposition to a variety of threats.
We couldn’t be extra excited that Sturdy Intelligence was named a Cool Vendor and stay up for taking AI safety to an unprecedented scale as a part of Cisco!
Gartner, Cool Distributors for AI Safety, Jeremy D’Hoinne, Bart Willemsen, Avivah Litan, Dennis Xu, 23 October 2024.
GARTNER is a registered trademark and repair mark and, COOL VENDORS are logos of Gartner, Inc. and/or its associates within the U.S. and internationally and are used herein with permission. All rights reserved. Gartner doesn’t endorse any vendor, services or products depicted in its analysis publications and doesn’t advise expertise customers to pick solely these distributors with the best scores or different designation. Gartner analysis publications encompass the opinions of Gartner’s analysis group and shouldn’t be construed as statements of reality. Gartner disclaims all warranties, expressed or implied, with respect to this analysis, together with any warranties of merchantability or health for a specific function.
We’d love to listen to what you suppose. Ask a Query, Remark Beneath, and Keep Linked with Cisco Safe on social!
Cisco Safety Social Channels
Share: