AI versus Identity Attacks: Deep Learning Software Bolsters Digital Fingerprinting
NVIDIA believes their Deep Learning software can thwart one of the hardest cyber threats to detect — identity attacks. The deep learning software bolsters digital fingerprinting workflows to detect and prevent identity attacks from wrecking website credential databases. According to research, stolen credentials make a lion’s share of all data breaches. Web applications are the worst hit in credential abuse. In the modern era of zero trust frameworks, companies need to think differently to detect any attack masquerading as a legitimate action. Artificial Intelligence versus identity attacks is definitely the clash of the titans that seemingly could decide the future of cybersecurity in the IT industry. NVIDIA’s deep learning software is an important arsenal in the hands of security experts in this battle.
The director of cybersecurity engineering and R&D at NVIDIA, Bartley Richardson, said – “We need to look for when Bartley is not acting like Bartley.” Digital fingerprinting with deep learning software was proposed as a potent solution against any kind of identity attack targeting human data. This deep learning model was built for data-centric platforms such as user account, server, application and device on the network. The ML model would detect deviations in behaviors and actions before alerting the security personnel about an attack.
The models would learn individual behavior patterns and alert security staff when an account was acting in an uncharacteristic way. That’s how they would deter attacks.
Why Big Data is both the problem and a solution?
Companies collect and analyze Big Data-sized information on their network events every day. It’s only a fraction of the big data that companies could log if they had the resources. According to Daniel Rohrer, NVIDIA’s vice president of software product security, the security problems arise from big data characteristics. The fact that it’s a big-data problem is also good news, Rohrer said in a talk at GTC in September (watch free with registration). “We’re already well on the way to combining our cybersecurity and AI efforts,” he said.
NVIDIA tested thousands of AI models in tandem on NVIDIA Morpheus, an AI security software library announced a year earlier. It took the team two months to build a proof of concept and another two months optimizing each portion. Nearly 50 NVIDIANs reviewed this work. They mostly comprised of people from the security operations and product security teams, and IT folks who would be alpha users. By October 2022, NVIDIA was able to develop a powerful AI-powered Deep Learning software for digital fingerprinting. It was similar to a LEGO kit that anyone from cybersecurity domain could use to create customized security frameworks against identity attacks.
Three months later, in early October, they had a solution NVIDIA could deploy on its global networks — security software for AI-powered digital fingerprinting.
The software is a kind of LEGO kit, an AI framework anyone can use to create a custom cybersecurity solution.
Deep Learning Software: Tested and Released
NVIDIA has introduced the latest deep learning capabilities for digital fingerprinting AI workflow. It is included with NVIDIA AI Enterprise 3.0 announced in December.
For identity attackers, “the models Bartley’s team built have anomaly scores that are off the charts, and we’re able to visualize events so we can see things in new ways,” said Jason Recla, NVIDIA’s senior director of information security. This step substantially cut down on the incidents that require IT team’s intervention. For instance, now IT teams investigate only 8-10 incidents against 100 million network invents in a week! Therefore, deep learning software was able to detect a potent threat within minutes instead of weeks.
“Our software works well on major identity attacks, but it’s not every day you have an incident like that,” Richardson said. “So, now we’re tuning it with other models to make it more applicable to everyday vanilla security incidents.”
Meanwhile, Richardson’s team used the software to create a proof of concept for a large consulting firm.
“They wanted it to handle a million records in a tenth of a second. We did it in a millionth of a second, so they’re fully on board,” Richardson said.
Recommended: Promising Success: 5 AI Trends for Business Owners and CEOs
The Outlook for AI Security using Deep Learning
NVIDIA is now focusing on developing accelerated computing platform for AI implementation. This would enable IT teams to secure digital assets and identities using credible DL-trained data pipelines. Richardson has postulated passwords and multi-factor authentication will be replaced by models that know how the user acts and behaves with data. For example, how fast a person types, with how many typos, what services they use and when they use them. This level of profiling with detailed digital identities will prevent attackers from hijacking accounts and pretending they are legitimate users.
“Data on network events is gold for building AI models that harden networks, but no one wants to share details of real users and break-ins. Synthetic data, generated by a variant of digital fingerprinting, could fill the gap, letting users create what they need to fit their use case.” – NVIDIA’s Nicola Sessions
In the meantime, Recla has advice security managers can act on now.
“Get up to speed on AI,” he said. “Start investing in AI engineering and data science skills — that’s the biggest thing.”
Digital fingerprinting is not a panacea. It’s one more brick in an ever-evolving digital wall that a community of security specialists is building against the next big attack.
You can try this AI-powered security workflow live on NVIDIA LaunchPad starting Jan. 23. And you can watch the video below to learn more about digital fingerprinting.
Comments are closed.