According to the Pew Research Center, 68% of technology innovators, developers and business leaders expect ethical principles focused on the public good to continue to be ignored in most AI systems until in 2030.
As AI works to match human capabilities, a major concern is that it could potentially overwhelm our ability to control it within an ethical framework. As a result, there is a growing movement to create ethical guidelines for AI systems. But to enforce the ethics of AI, the industry must first define those ethics.
Different people and organizations have attempted to create ethical codes for AI over the years. For example, in 2016 the EU adopted the GDPR, which laid the foundation for a model on how to apply ethics related to intangible tools that impact human behavior. This has forced companies to consider the ethics of using and storing personal information, a crucial first step when it comes to AI.
Yet today, there is no widely accepted ethical framework for AI, nor a way to enforce it. Obviously, ethical AI is a big topic, so in this article, I’d like to narrow it down and look at it through the lens of network monitoring technologies.
See also: Top AI software
AI and network monitoring
AI has many potential benefits when applied to network monitoring and performance. While many staff fear being replaced by AI, in the networking space, the growth of AI actually signals improvement, not displacement.
In fact, AI in IT surveillance environments can streamline complex networks, automate specific tasks, and help increase the efficiency of detecting and resolving threats, to name a few areas. It can also simplify IT’s role in monitoring and help find the root cause of issues faster.
Let’s look at some specific examples of AI in network monitoring, so that we can better understand the key ethical issues later.
- Anomaly detection uses AI/ML to understand normal versus abnormal behavior (to establish baselines) on a network. It is used to create models of what typical traffic looks like tailored to specific locations, users, and time aspects. These models can be very detailed, down to the specific application. They allow organizations to understand patterns by extracting application functionality from a network perspective.
- Predictive analytics leverages data with AI/ML to predict potential issues that may arise on a network in the future. Like anomaly detection, it also uses data analysis to learn about historical patterns and events, and finds and learns patterns that may be causing problems.
- Automating also uses AI/ML to determine what could be the root cause of a network problem and fix it automatically. ML techniques such as decision trees or more sophisticated techniques can create learned processes for diagnosing problems rather than creating rule-based manual systems that can be error-prone and difficult to maintain.
See also: Data Analysis Trends
Networking and ethical issues related to AI
While AI can provide a new level of visibility and problem solving when applied to network monitoring, there are also ethical considerations or questions that the industry should consider or ask. There is a lot of debate around ethical AI, but most agree that AI ethics is a system of moral principles and techniques intended to inform the development and responsible use of technologies. of AI.
But what does this mean in the network monitoring space? I don’t claim to have all the answers, but I do have some key questions that we should all ask ourselves and work together to answer.
- Is the data used in accordance with applicable privacy and protection regulations, be it GDPR in the EU or other regulations? Network Data may contain personal, behavioral and trending information. It’s important to make sure it meets regulations, especially as AI/ML systems ingest more data.
- Does the data have potential for bias when features are extracted and used to train models? As models are developed, humans bias detections based on patterns that can be correlated to gender, race, ethnicity, etc. This is more pronounced with social data, but users generating network traffic may have patterns specific to a cohort group. While this does not create social bias, it could create patterns that may not work universally as intended.
- Are the actions recommended or taken based on the analysis and the potential implications? As observed with self-driving cars, there are always “corner” cases or unseen scenarios that AI systems may not have been trained on. Exploration of all possible outcomes, even if not supported by data, should be considered and justified.
It is important to note that the industry is not completely starting at square one, but it is only in its infancy for AI standards. Today, there are IT initiatives designed to help create and shape ethical AI. These include at a general level the GDPR, which does not deal directly with the ethics of AI, but it does deal with data protection and privacy, which has implications for the use of that data. for AI.
There is also a proposed EU AI law that will specifically address rules around the development and use of AI-based products. But most of the time, AI ethics are left to tech developers at this point – something that needs to change in the future.
As AI innovation continues, having safeguards and standards in place will be essential. Unchecked AI is universally seen as a recipe for disaster.
But AI produced and implemented with ethical guidelines has incredible potential in the network monitoring space to save NetOps teams a lot of time and resources when it comes to collecting, analyzing, to design and secure networks.
See also: What is data visualization?
About the Author:
John Smith, CTO and co-founder of LiveAction