Identify the Problem to Be Solved
Ken Xie, who was born in China, is one of the most successful entrepreneurs in Silicon Valley. His focus has been on transforming cybersecurity technology, such as by developing the first ASIC-based firewall and VPM appliance in the 1990s.
- PDF / 271,804 Bytes
- 24 Pages / 439.37 x 666.142 pts Page_size
- 62 Downloads / 199 Views
3 Identify the Problem to Be Solved Where do you start with your AI project? Ken Xie, who was born in China, is one of the most successful entrepreneurs in Silicon Valley. His focus has been on transforming cybersecurity technology, such as by developing the first ASIC-based firewall and VPM appliance in the 1990s. He also started several breakout companies. There was NetScreen, which was sold to Cisco for $4 billion in 2004. Then he co-founded Fortinet, which has seen consistently strong growth (the other co-founder was his brother, Michael). The market value is about $22.5 billion and there are more than 450,000 customers across the globe. Then what about AI? What has the company done with this technology? It has definitely been a major priority. In fact, AI has been essential in not only improving cybersecurity systems but also in dealing with the talent shortage of skilled technical employees.
© Tom Taulli 2021 T. Taulli, Implementing AI Systems, https://doi.org/10.1007/978-1-4842-6385-3_3
64
Chapter 3 | Identify the Problem to Be Solved “The cybersecurity skills gap is very real and hasn’t shown signs of improving in the foreseeable future,” said Derek Manky, who is the Chief of Security Insights and Global Threat Alliances at FortiGuard Labs, which is a division of Fortinet. “At the same time, the threat landscape is becoming more complex, posing significant risk for just about any organization. For example, and specific to AI, in recent years, the security industry has seen cybercriminals investing in weaponizing AI for their own nefarious activities, developing AI-based attacks. For those organizations that have the right technology in place, AI has emerged as the ultimate tool for cybersecurity professionals to close the skills gap and ‘fight fire with fire’ against prolific hackers already using AI technology.” To this end, Fortinet has invested heavily in creating self-learning deep neural networks to essentially act like a virtual security analyst that investigates attacks. At the heart of this is a patent-pending algorithm that is based on more than 20 million clean and malicious files, and has classified the threats into more than 20 attack scenarios. “When leveraging Fortinet’s AI technology, what would have traditionally taken a cybersecurity expert a few days to narrow down, AI can detect in minutes or sub-seconds,” said Manky. For Fortinet, AI is an ongoing process and requires a strong commitment of resources. “Learning is a long game, and is always evolving and the main task— both for internal processes, pipelines, and customer protection—is always reducing the workload and working smarter,” said Manky. “Machine learning is specific to a task. The largest model we have is filtering malicious codeblocks from code, which we started supervised training about over five years ago. We have billions of features mined from our 800 million malware samples, as a result, making the model more precise and acceptable to deploy to production.” True, Fortinet has some inherent advantages when it comes t
Data Loading...
