cool hit counter The cyber offensive and defensive battles of 2018 must be aided by AI_Intefrankly

The cyber offensive and defensive battles of 2018 must be aided by AI


E Security Jan. 1 - The learning capabilities of artificial intelligence (AI) are predicted to be key to cyber defense and cyber attacks.

2016summer of 2009, Hackers with7 A hacked machine travels to Las Vegas, USA to participate in the US Department of Defense Advanced Research Projects Agency(DARPA) hosted“ Web Super Challenge”(CGC)。 Over dozens of rounds of competition, Each machine seeks to find more software vulnerabilities," attempt" put to good use, And fix it before other machines use the same tactics to win the game。 The participating teams have given these machines the processing power、 Software analysis algorithms and utilization tools。

The CGC is the only all-machine hacking competition to date, and the initial winner is a machine called Mayhem, which is now on display as the first "non-human entity" at the Smithsonian's National Museum of Natural History. This machine had won the highest honor a hacker could dream of, the Black Badge, at DEFCON. Mayhem took part in another showdown with hackers in August 2017 and lost to them in the end. While it can work night and day, it lacks the energy and motivation to compete with humans. Mayhem eventually lost to humans due to the machine's lack of creativity, intuition and decisiveness.

What's coming in 2018?

However, this is set to change in 2018. The development of computing power, theoretical and practical concepts of AI research, and cybersecurity breakthrough technologies will drive machine learning algorithms and techniques as a key component of cyber defense, even as much as for attacks. Hackers are improving the technology of machines and working with them to meet new competitive challenges, for example, the Shellphish team has launched the open source exploit automation tool "angr".

From a defense perspective, cybersecurity experts are already using a lot of automated and machine-driven analysis. On the other hand, automation capabilities are constantly being applied in offensive attacks. Cybersecurity firm Cylance found through a survey at Black Hat USA 2017 that 62% of information security experts believe hackers will weaponize AI and start using it to carry out attacks in 2018.

At DEFCON 2017, a data scientist at US endpoint security vendor Endgame demonstrated and released a malware operating environment for the OpenAI Gym, an open source toolkit for learning algorithms. Endgame has developed an automated tool to learn how to modify bytes in code to hide malicious files and evade detection by anti-virus engines.

With the development of such new tools, coupled with competition driving innovation, it's not hard to imagine that this development ladder will next give rise to automated systems that adapt, learn new environments and identify vulnerabilities, but such systems could also be played by hackers.

Original Article: https://www.easyaq.com/news/1556069732.shtml

E Security Note: This article is exclusively compiled by E Security, please contact for authorization to reprint, and retain the source and link, no deletion of content.


Recommended>>
1、Miners Adventures Game World Premiere Meet and Greet Strategic Signing Ceremony was a success
2、FlashCC2015 Installation Tutorial
3、Notes on the use of the visualization tool mataplotlib I x Commonly used charts
4、BMW X3 is going domestic Add soundproof glass and autopilot How far away is the cabbage X3
5、Creative programming courses from life to life

    已推荐到看一看 和朋友分享想法
    最多200字,当前共 发送

    已发送

    朋友将在看一看看到

    确定
    分享你的想法...
    取消

    分享想法到看一看

    确定
    最多200字,当前共

    发送中

    网络异常,请稍后重试

    微信扫一扫
    关注该公众号