Artificial intelligence (AI) has become a prominent industry buzzword. From messaging and chatbots, to sophisticated enterprise applications, it’s clear that AI is here to stay.
People continue to hear mentions of how AI will advance society, but the debate remains as to how AI can and should be applied in practicality and whether the promise is real.
Today, many organisations continue to fail to effectively apply AI to solve specific business cases. The hype around AI has led to a trend where every vendor claims to leverage it in their technology, solutions or products, causing extreme confusion, and in some cases, frustration among technology users.
Currently, investments in AI remain high. A study by McKinsey Global Institute reported that in 2016, companies invested between $26 billion and $39 billion in AI and the investments continue to soar. These investments and the organisations that boast AI capabilities will undoubtedly increase, but the high-level and widespread adoption in most industries will remain relatively low compared to more traditional solutions.
Contrary to industry hype, AI cannot and should not be used to solve every problem. In fact, we are even seeing companies use AI in situations where the capability isn’t even needed to improve the product or software.
To understand where AI should be used and will be most successful, one must understand what AI really is. AI, or machine learning, refers to a broad set of algorithms that can solve a specific set of problems, if trained properly. While integrating machine learning into a product is trivial, effectively training the algorithms to perform their task is not.
>See also: Ethical AI – the answer is clear
AI works best when large amounts of rich, big data are available. The more facets the data covers, the faster the algorithms can learn and fine-tune their predictive analyses. According to industry predictions, in 2018, AI’s greatest limitation — high quality data — will become more evident. Successful machine learning depends on large and broad data sets.
From a technical standpoint, machine learning or AI features have never been more accessible than they are today. In the cybersecurity industry, when effectively leveraged, AI has the potential to help business leaders shore up their resilience to all types of cyber attacks. AI’s main promise in this space is the ability to consistently detect new and unknown threats in the absence of traditional indicators of compromise – such as a known pieces of malware.
With sufficient quality data available, AI techniques easily outperform traditional, signature-based approaches, which retroactively seek out the artifacts an attacker leaves during a breach.
Artificial intelligence can drive the creation of indicators of attack (IoAs), which can identify active attacks based on how an adversary behaves in the system, allowing organisations to prevent breaches. What’s even more important, the algorithms will learn from those behaviors for future-proofed protection.
Scale is another critical component of getting AI applied effectively. In order to swiftly and accurately analyse billions of security events in real-time, a capability that is needed for effective threat protection, the algorithms require a level of computational power and scalability. In most cases, that degree of enablement cannot be accomplished using old-school, on-premise architecture and conventional database methods.
Instead, technology companies turn to the cloud to secure the needed scalability, computational power, and resources. In the cyber security industry, the cloud enables the crowdsourcing of security data and threat intelligence to deliver instant protection against incoming threats to the entire customer community, regardless of its size.
Moving forward, businesses and buyers must carefully evaluate those organisations touting their AI capabilities, keeping a keen eye to ensuring the technology leverages the right data and capabilities to be truly effective. In the next wave of AI empowerment, the algorithms are commoditised, but whoever owns the data is king.
Sourced by George Kurtz, CEO and co-founder of CrowdStrike – a provider of next-generation endpoint protection, threat intelligence, and services