No Automated Agent Defines Threats Automatically
By Enterprise Security Magazine | Friday, April 12, 2019
Today, it appears that almost all cybersecurity products use artificial intelligence (AI). It is probably noticed that some experts are talking about the idea of a breakthrough while others are still fuelling the hype. The fact is, nearly all security experts see much potential of AI to solve current cybersecurity space challenges. Now, on the other hand, hackers are also taking advantage of AI's new methods, and the critical question at this point is, how significant the potential and impact of this new technology is?
There is actually no known application of AI in cybersecurity. No autonomous agent defines threats automatically. ML and AI certainly play a role in cybersecurity. For instance, they are very good at analyzing a lot of information and understanding what is normal and what is anomalous.
ML is also used to automate responses to common vulnerabilities and eliminate some of the time-consuming protocol heavy lifting. Some AI/ML-based systems have already proved successful in addressing complex tasks. Both these technologies can also be used to augment security staff.
Companies are zeroing in on vendors with an expansive approach to accommodating various analytics data sources. Businesses should always be careful about bold claims. If there is a pop-up like 'we detect unknown attacks automatically,' the chances are it’s ‘nonsense’. Most importantly, before buying any new solution, define the particular problem that is present and then determine whether ML or AI is the right way to solve the problem. Even a much better traditional way to solve the problem may exist.
Ultimately, organizations need to spend time in shaping the machine learning output with business context to make the results more meaningful and insightful. This enables analysts to spend time on the system and use its context and insights to infuse it.
Check this out: Top Cyber Security Companies