Resistant AI, a safety firm that helps to guard AI methods from focused manipulation, adversarial machine studying assaults and superior fraud, has raised $2.75 million in funding.
The spherical was co-led by Index Ventures (Jan Hammer) and Credo Ventures (Ondrej Bartos and Vladislav Jez) with participation of Seedcamp. Daniel Dines, CEO of UiPath, Michal Pechoucek, CTO of Avast and different angel traders have additionally invested. Ondrej Bartos joins the board of administrators on behalf of the traders.
With the elevated reliance on AI, malicious actors are actively exploring vulnerabilities which, due to the very nature of AI, are a lot tougher to detect. Working with clients in monetary companies, fintech and e-commerce, Resistant AI ensures that the AI-driven on-line and automatic processes that these corporations depend on, haven’t been compromised.
Martin Rehak, Founder & CEO, Resistant AI: “Traditionally, all methods that make high-value monetary selections change into focused. That is already taking place with the automated methods deployed by our fintech and monetary clients and we’re right here to guard them.”
Resistant AI is deployed as an extra safety on high of AI methods, and:
• protects credit score threat scoring, cost methods, fraud detection, anti-money laundering and buyer onboarding methods;
• detects solid paperwork submitted to mislead or manipulate automated processes;
• discovers relationships between seemingly unrelated transactions to establish superior fraud designed particularly to evade the present fraud prevention methods, corresponding to artificial identities, bust-out fraud, approval boundary probing and lots of others;
• defends in opposition to strategic assaults the place the attackers purpose to repeat the underlying model or the delicate knowledge used to coach the model.
Michal Smida, CEO of Twisto funds, described the ResistantAI service as follows: “Resistant AI is a critical part of our security infrastructure, giving us the confidence that we can rely on our machine-learning-based credit scoring system, which we use to automate the approval of credit payments. And it is because of Resistant AI that we don’t need to put in excessive security measures and make the customer experience much better.”
Jan Hammer, Common Associate at Index Ventures added: “Automation, effectivity and reliability are cornerstones of monetary innovation. As machine studying takes increasingly more nuanced monetary selections, it must be protected. And this isn’t true solely in finance, however the assaults will quickly unfold to different domains as effectively. Dangerous actors are able to make the most of each vulnerability. Extra exercise is going down on-line right now than ever earlier than, a pattern additional accelerated by Covid-19. The necessity for options corresponding to these from ResistantAI has by no means been higher.”
Ondrej Bartos, Common Associate of Credo Ventures, who joins the Resistant AI board of administrators with this spherical commented: “We at Credo are all the time delighted to again founders of our previous profitable portfolio corporations. Due to this fact we’re excited to co-lead the funding into Resistant AI whose founders we’ve labored with carefully up to now years and in our view have confirmed deep expertise experience and foresight which is strictly the toolset they should deal with safety of AI selections.”
Resistant AI was based in 2019, and is made up of a crew with deep technological background in machine studying, AI and community safety. The founders have greater than 12 years of expertise making use of AI within the safety area. Their prior startup, Cognitive Safety, based in 2009 as a spin-off from Czech Technical College and funded by Credo Ventures, was acquired by Cisco Methods in 2013 and have become the premise for Cisco’s R&D heart.