Generative artificial intelligence (AI) represents a new frontier for transformative productivity.
With over 300,000 customers worldwide harnessing our data-driven solutions, SolarWinds is well-positioned to leverage AI to enrich the lives of the IT professionals we serve. But for this exciting technology to yield real value over time, it’s crucial to build it sustainably.
Since 2021, our Secure by Design principles have guided our security and cyber resilience approach. Now, we’re extending this model to establish optimal practices around artificial intelligence. AI by Design is our approach to AI development, consisting of four foundational principles. They create a framework to help our customers establish a secure, productive, and enduring relationship with our AI-driven solutions. In a series of articles to appear on Orange Matter in the coming weeks, we’ll explore all four AI by Design principles in detail. But for now, let’s take a brief look at each:
Principle 1: Privacy and Security
At SolarWinds, innovation means secure innovation. AI by Design’s Privacy and Security principle dictates personal data collection, storage, use, and protection. This includes regulating access through role-based access control (RBAC), multi-factor authentication (MFA), and privileged access management (PAM) to prevent unauthorized entry. Anonymization and pseudonymization techniques help mitigate privacy risks by removing or replacing identifiable information. Robust auditing and monitoring mechanisms enable the detection of unauthorized access and facilitate timely responses to security breaches. These are just some of the approaches that SolarWinds is actively implementing to achieve maximum safety regarding data and AI.
Principle 2: Accountability and Fairness
Without the correct precautions, AI systems can perpetuate biases present in the data used to train them, leading to discriminatory outcomes. AI by Design’s Accountability and Fairness principle ensures that there is a human in the loop to regulate the decisions made by AI. Feedback and validation mechanisms are built in so that negative user experiences are properly recorded and addressed. For example, our Agent Assist feature with Gen-AI allows the human agent to edit AI-generated responses before sending them to the requester. Feedback on AI suggestions is regularly collected and reviewed to improve model behavior over time.
Principle 3: Transparency and Trust
Decisions must be accompanied by explanations to build a strong working relationship with AI systems. This transparency fosters a deeper confidence in the outcomes of AI-driven processes. A well-defined machine learning operations (MLOps) pipeline facilitates the demonstration of the rationale behind AI-generated actions, enabling continuous refinement based on evolving needs. Transparency entails demystifying the processes behind AI decisions and outputs. This creates trust, arguably the most critical ingredient for fostering a positive response to new technologies.
Principle 4: Simplicity and Accessibility
For IT and DevOps teams managing mission-critical applications, user experience around AI is pivotal. A seamless, intuitive experience prevents fatigue and streamlines decision-making. This ease of use is invaluable for teams processing vast data streams from diverse workloads. A key aspect of manifesting simplicity in AI-driven experiences is to be mindful of the procedures that our users are accustomed to. Rather than introducing brand-new experiences (requiring training and causing additional cognitive load), we identify familiar patterns and build on them. One example is our anomaly-based alerts, designed to reduce the fatigue caused by alert storms. By taking existing behavior and practices as a starting point, we help ease the transition to this new technology.
AI by Design: An Evolving Framework
Given the exciting prospects of artificial intelligence, it's easy to be hasty in development and integration. This can lead to unforeseen consequences. AI by Design takes in expert knowledge from both private and public spheres so that we can be wise in our approach to AI. AI by Design is crafted to ensure that our customers can leverage the very best AI has to offer while minimizing potential downsides. Crucially, it isn’t a static framework but a dynamic and evolving set of guidelines that will adapt to emerging insights.
In the coming weeks, we'll explore each AI by Design principle in detail on OrangeMatter, giving you a deeper understanding of how these principles shape our solutions.
