HiddenLayer raises $50M to defend enterprise AI models
Head over to our on-demand library to view sessions from VB Transform 2023. Register Here
HiddenLayer, an Austin, Texas-based cybersecurity startup born out of a cyberattack that exploited machine learning code at the founders’ prior company, has announced a $50 million Series A funding round today to further harden the defenses of the rapidly growing number of AI models being adopted by enterprises.
The round was led by M12, Microsoft’s Venture Fund, and Moore Strategic Ventures, with participation from Booz Allen Ventures, IBM Ventures, Capital One Ventures, and Ten Eleven Ventures.
“AI’s unapparelled rate of adoption fuels us to move even faster in achieving our mission to give every security professional the right tools and expertise for embracing AI securely,” said Chris Sestito, CEO and Co-Founder at HiddenLayer, in a statement in the company’s press release announcing the round.
Already, HiddenLayer helps safeguard AI/ML models used by a number of Fortune 100 firms across sectors inclucing finance, government and defense, and cybersecurity.
Event
VB Transform 2023 On-Demand
Did you miss a session from VB Transform 2023? Register to access the on-demand library for all of our featured sessions.
What HiddenLayer does
As previously covered by VentureBeat last year following its emergence from stealth, HiddenLayer has built a number of tools as part of its “MLSec” Platform for safeguarding enterprise machine learning (ML) and AI models.
These tools don’t actually access the models, nor compromise the proprietary data and technology of clients. Instead, the passively monitor the performance and operations of enterprises ML/AI models and linked applications in realtime, scanning overarching vulnerabilities and offering recommendations for hardening them, as well as detecting injection of malicious code/malware and deploying defense mechanisms to cut off the attackers and isolate any intrusions.
HiddenLayer’s MLSec Platform ships with a simple but powerful dashboard allowing security managers to get access to all the information they need about the security state of their enterprise ML/AI models at a glance. It also automatically lists security issues and alerts in order of priority depending on the severity of the issue, and stores data for the compliance, auditing and reporting that a business may be asked to do.
HiddenLayer further offers consulting services from its team of Adversarial Machine Learning (AML) experts who stay atop the latest trends in security and the newest threats. They can perform threat assessments, training for a client’s cybersecurity and dev ops personnel, and perform “red team” exercises to ensure the client’s defenses are working as intended.
Influential partner
Earlier this year, the company struck a partnership with white-hot enterprise data lakehouse provider Databricks, allowing Databricks enterprise customers to use HiddenLayer’s MLSec Platform directly on their models running on Databricks’ lakehouses.
“The integration is model agnostic and includes model scanning and model detection and response,” explained HiddenLayer at the time in a blog post announcing the partnership. “This enables Data Scientists and ML Engineers to add security to their models with no code or behavioral changes to their environment. As the model is loaded, it will be scanned by HiddenLayer’s model scanner to ensure integrity as well as security. If an attack is detected, the integration will handle the response accordingly without any human interaction needed.”
What’s next for HiddenLayer’s quest to secure enterprise AI?
HiddenLayer was founded after co-founders Sestito (CEO), Tanner Burns (chief scientist) and Jim Ballard (chief information officer) after the three encountered a cyberattack on ML models at the prior company, Cylance, a security startup.
As recollected on HiddenLayer’s website, the incident occurred when “attackers had exploited Cylance’s Windows executable ML model using an inference attack, exposing its weaknesses and allowing them to produce binary files that could successfully evade detection and infect every Cylance customer.”
While worrisome and stressful at the time, the trio realized then that attacks on ML/AI would only increase in the near future as more enterprises sought to adopt generative AI into their workflows due to the technology’s great promise at increasing efficiency and performance.
Today, HiddenLayer is growing rapidly, having quadrupled its headcount in the last year. Now flush with its Series A cash, it plans to hire another 40 personnel by the year’s end, and cont continue growing its client base.
VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.