SHANGHAI, April 18, 2018 /PRNewswire/ -- KEEN has invited global security experts to participate GeekPwn 2018, its worldwide security geek contest for "PWNing" devices and systems to bring about a safer smart life. The GeekPwn 2018 contest will be held in Las Vegas in August (around the same time as top contest DEF CON) with a carnival session being held in Shanghai in October. Maximum individual prize awarded will be $150,000 for each PWN competition, of a total pool of 5 million RMB ($800,000).
GeekPwn 2018 will include Special Challenges and PWN Everything Challenges. The Special Challenges, featuring different scenarios in fields such as security, AI and machinery, include CAAD or Competition on Adversarial Attacks and Defenses (Las Vegas), Robot Agent Challenge (Las Vegas/Shanghai) and the Hacker Room Challenge (Shanghai). PWN Everything, which encourages contestants to find security vulnerabilities and creative PWN methods, is divided into Vulnerability based PWN (Shanghai) and Non-Vulnerability based PWN (Las Vegas/Shanghai).
GeekPwn 2018 will create a global platform for cyber security researchers and experts, with technical advantages and diversified and out-of-box mechanisms. The new setup of challenges and rules are going to open wider horizons for security development and establish a security contest benchmark globally.
CAAD makes AI more Secure
GeekPwn will organize CAAD (Competition on Adversarial Attacks and Defenses) in Las Vegas in August. CAAD is focusing on sample studies of machine learning or AI confrontation between "attack" algorithm and "defense" algorithm. What is learned from the confrontation could potentially be adopted in the battlefields of cyber warfare in future. GeekPwn is honored to have top AI experts as CAAD judge committee members, including Ian Goodfellow and Alexey Kurakin from Google Brain and professor Dawn Song from UC Berkeley.
Though it is important to train AI algorithms, the potential threats of AI can't be ignored, which manifest themselves most seriously in the fields of cybercrime, physical attacks and political interference. GeekPwn, the first worldwide security geek contest covering AI security, also focuses on these AI security issues.
Each of the three CAAD sub-competitions has a maximum prize of $30,000 USD.
Robot Agent and Hacker Room Challenge
Besides PWNing traditional devices, robots have joined the battlefield in GeekPwn for the first time. In the Robot Agent Challenge in Las Vegas, each team is required to make an "Agent" robot with arms, which can "sneak into" an office and hack different physical facilities. Contestants will control the robots to bypass laser intrusion detectors and surveillance cameras, leave a covert listening device, open a safe by a entering password using a keyboard, boot target computers from USB devices, and so on. The maximum prize awarded in the challenge will be $30,000.
Another highlight of GeekPwn 2018 is the Hacker Room Challenge. This innovative contest showcases the whole process of "attack and defense" in the hacking world in a stereoscopic and proactive way, which makes GeekPwn 2018 the year's most exciting and interesting contest or "GeekGame".
PWN Everything on Vulnerabilities and Methods
To better protect people's smart life in the digital age and ensure AI's development in a sustainable way, GeekPwn will continuously accept PWNs on a wide variety of targets. Contestants are encouraged to find security vulnerabilities of target systems or try new creative PWN methods. Smart devices, AI products, libraries, frameworks and IoT products that are commercially available or widely used are all acceptable PWN targets.
In Vulnerability-based PWN (Shanghai), attackers can get system control, access private data or break through original security mechanisms by exploiting security vulnerabilities. For example, contestants obtain the admin rights of a facial recognition door by using logic errors or control others' drones by simulating the control app.
In the Non-Vulnerability based PWN (Las Vegas/Shanghai), the attack is not exploiting target systems' vulnerabilities, but using new creative methods to PWN. For example, an attacker uses machine learning to study a specific human voice and simulates it to get past a voice recognition system.
The maximum individual prize awarded will be $150,000 USD for each PWN, the largest prize of all GeekPwn 2018 contests.
As one of the world's leading platforms for cybersecurity researchers, GeekPwn enables security researchers and executives around the world to share their thoughts and findings. Since 2014, GeekPwn has successfully held 8 sessions in Beijing, Shanghai, Macau, Hong Kong and Silicon Valley, and responsibly disclosed hundreds of critical security vulnerabilities and awarded millions in USD to contestants.
KEEN Cloud Tech endeavors to help the world's leading software manufactures who have adopted advanced cybersecurity engineering methodologies to discover and fix cybersecurity vulnerabilities. Over the years, KEEN has discovered and reported hundreds of high-priority vulnerabilities to leading companies such as Microsoft, Apple, and Google. KEEN, as the 1st Pwn2Own champion team in Asia and five-time Pwn2Own winners within a span of 3 years, has achieved a sound reputation through its achievements and innovative contributions. GeekPwn Lab is the cybersecurity research team of KEEN Cloud Tech. The team is dedicated to cybersecurity research about cutting-edge and future technology applications, including infrastructure, internet protocol, mobile payment, IoT and Artificial Intelligence.