AI vs. Algorithms
Algorithms: they have become something of a “can’t live with them, can’t live without them,” aspect of modern life. They impinge on our daily lives by attempting to figure out the kinds of ads we should be served – I currently get ads about women’s clothing, golf (which I don’t play), and other non-related items.
Perhaps one of the biggest algorithmic annoyances is airline prices. They are set up to automatically adjust to maximize the revenue for airlines. They are good at it, almost too good. You make a quick search of airline prices one day as a bit of a homework. A couple of days later you check again and the price has shot up. The cause could be a cookie that noticed your return and triggered a pricing algorithm. Or lots of people searching for that flight triggering a price hike.
You see the same thing at some theme parks. One in Florida charges more based on peak times and demand. Super Bowl weekend hotel rates are another example.
Things are only going to get worse – or better – once artificial intelligence gets its act together and does a better job of fine tuning algorithms to better gauge human behavior, emotions, and of course, set the best possible price for everything humans want.
Let’s take a look at how AI and algorithms are tied together.
Also see: Top AI Software
What is an Algorithm?
The Encyclopedia Britannica definition of an algorithm is simple and to the point: “An algorithm is a specific procedure for solving a well-defined computational problem.”
In essence, an algorithm is a series of steps that are followed to solve a mathematical problem or to complete a computer process. Their development has been fundamental to such areas as AI, databases, graphics, networking, operating systems, and security. Algorithms go beyond computer programming as they require understanding of the various possibilities available when solving a problem.
Also see: The Future of Artificial Intelligence
What is Artificial Intelligence?
Encyclopedia Britannica defines artificial intelligence as, “The ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings. The term is frequently applied to the project of developing systems endowed with the intellectual processes characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from past experience.”
Massive amounts of work have been done over many decades to bring AI up to speed. Computers can be programmed to carry out complex tasks. They can even play chess well enough to beat a grand master, at times.
But that is based on a very limited 8 by 8 square board over two dimensions. Despite what have been regarded as colossal advances in processing speeds and memory, AI can’t match the flexibility of humans across a wider zone of activity. But when narrowed down to very specific areas, AI has led to multiple gains in areas such as search engines, handwriting recognition, and even some kinds of medical diagnosis.
Also see: The History of Artificial Intelligence
How Does AI Use Algorithms?
Algorithms can be regarded as the essential building blocks that make up artificial intelligence. AI can use various algorithms that act in tandem to find a signal among the noise of data and find paths to solutions at levels of complexity at which humans would not be capable. AI makes use of computer algorithms to impart autonomy to the data model and emulate human cognition and understanding.
Algorithms enhance AI by making the system smarter. They are used for calculation, data processing, and automated reasoning. Algorithms enabled by AI include natural language processing (NLP), computer vision, and facial recognition.
Using machine learning algorithms, for example, AI software can analyze, detect, and alert anomalies within a network infrastructure. Some of these algorithms attempt to mimic human intuition in applications such as the prevention and mitigation of cyber threats. Done well, this helps to alleviate the burden from understaffed cybersecurity teams.
In some cases, algorithms are designed so well that they can be used to automatically take remediation steps following a breach. But in many other cases, they still rely on the human touch to determine what is really going on.
“Algorithms must be developed to deliver artificial intelligence,” said Rick Wagner, Senior Director, Product Management, SailPoint. “A key element to the algorithms is data that can be provided for analysis. The relationship of the data to the problem and the depth/breadth of the data is crucial to ensure the greatest degree of accuracy that can be delivered by AI.”
Algorithms, too, are the backbone of machine and deep learning. It is the algorithm that is the substitute for the human processing the information. Perhaps most important, algorithms enable analysis of all aspects of data and summarization to key stakeholders that will improve the speed and quality of security related issues.
“A machine processes algorithms with the data that is provided,” said Wagner. “The quality and relevant quantity of the data provided to the algorithm and the completeness of the scope of the algorithm ultimately determines the quality of the artificial intelligence.”
Also see: Best Machine Learning Platforms
Algorithms and AI in Security
Developers of security technologies are among the firm believers in the value of algorithms and AI. Security information and event management (SIEM) systems utilize them to sift through vast numbers of computerized logs and alerts. They highlight patterns and anomalies that might otherwise be missed.
Endpoint detection and response systems (EDR) use AI and algorithms to monitor ports, and other means of entry into computer systems. Threat detections systems, too, use them to spot the first signs of an incursion. User and Entity Behavior Analytics is another largely AI-driven security field that harnesses advanced algorithms. Yet another AI-driven security approach is detecting zero-day attacks or “unknown unknowns,” which cannot be achieved without AI automating baseline profiling and anomaly detection.
Finally, AI automation helps security response teams by ranking security alerts, reducing alert fatigue, and suggesting corrective actions. These approaches are taking the world of cybersecurity to another level – and it’s about time. Cybersecurity is already playing catch up with the increased sophistication of cybercriminals.
AI has that non-human capability of being able to find a needle in a haystack of data in a very short time frame. With the threat posed by incursions and ransomware, security technology needs to be able to detect and respond as close to real time as possible. But the amounts of data to be analyzed lies far outside the range of rapid inspection by a human pro in IT.
“AI and analytics are the key enablers of enhanced security,” said Stanislav Miskovic, Vice President of AI, Gluware, a company that is building a platform which will leverage and unify data across the networking stack. “Security needs to cover a much wider footprint today, which cannot be done without the help of AI. The number of attack surfaces is too great and data volume too huge for inspection without AI assistance.”
Also see: Top Digital Transformation Companies
Algorithms and AI in HR
HR is another sector that is now making heavy use of algorithms, usually in conjunction with AI. The global human resource technology market spending exceeded $24 billion in 2021. It is predicted to rise to $36 billion by 2028. These days, HR spending is heavily weighted toward tools and platforms that offer AI capabilities, are cloud-based, and come with built-in security.
Case in point: AI algorithms are being integrated into recruitment tools. Applicant tracking, social sensing, sentiment analysis, and hiring platforms are all utilizing algorithms, AI, and analytics to measure enterprise and employee productivity and morale. Modern natural language processing (NLP) techniques, for example, can accurately assess in real time the sentiment, toxicity, and hot topics of conversation occurring in a workplace.
“NLP tools give leaders unprecedented real-time insight into an organization’s collaboration environment,” said Jason Morgan, Vice President of Behavioral Intelligence, Aware. “Tracking organization sentiment and toxicity can arguably serve as key performance indicators for leaders who are supporting company culture, engagement, and wellbeing.”
AI and Algorithms: People Come First
This coming together of algorithms and AI, coupled with automation, is a vital part of fields like analytics, security, and HR. Some experts expect that technology will replace humans entirely. But that is a mistake. The key value of AI and algorithms is in data visibility and rapidly assessing mountains of data to find patterns. Decision making can be set up on a limited basis. But there is no replacing human judgment, even as flawed as it can be.
“When AI is being applied to make a decision on behalf of an individual or, to suggest a decision for an individual, the user must have the ability to understand how the decision/suggestion was derived,” said Wagner. “The human decision maker should always have the ability to override the decision/suggestion of the machine as there are elements that the machine cannot take into consideration where the human can.”
However, there may be some areas where bypassing human decision making might be desirable. James Katz, Boston University College of Communication’s Feld Professor of Emerging Media and co-editor of Journalism and Truth in an Age of Social Media, noted that of the many different roles offered up for replacement by computer algorithms, journalists were the most likely to be “voted off the island.”
“Most people want their news without a coating of bias by reporters,” Katz said. “Sadly, there is widespread and warranted skepticism about the press’s conduct in this regard. To me, the survey results suggest that many people, justifiably or not, expect that computer-algorithms could be more objective news providers than human reporters.”
But those in the news industry should not despair, added Michelle Amazeen, associate professor at Boston University’s College of Communication and director of the college’s Communication Research Center.
“Some journalists, such as fact-checkers, are steadfastly pursuing AI to help them do their jobs more efficiently,” Amazeen said. “Moreover, research has shown that when journalists do their work in tandem with algorithms, audience perceptions of bias are attenuated. These findings underscore the promise of AI in the journalism industry.”
Also see: What is Artificial Intelligence
The post AI vs. Algorithms appeared first on eWEEK.