Why responsible AI is a business imperative

The introduction of generative artificial intelligence (AI) has sparked debates about potential AI-related harms, ranging from concerns about training data to military applications to the ethical practices of AI vendors. While critics of AI, known as “AI doomers,” have called for regulations to address long-term existential risks, they have also raised issues about current legislation focusing on competition, access, and consumer protection.

On the other hand, major AI companies like Microsoft and Google are now releasing annual transparency reports detailing how they develop and test their AI services. These reports emphasize a shared responsibility model for enterprise customers using their tools and services, similar to the approach taken in cloud security. This becomes particularly crucial as “agentic” AI tools capable of autonomous actions become more prevalent.

As AI systems, both generative and traditional, are already in use within organizations and customer-facing tools, the responsibility of AI governance needs to shift from data science teams to Chief Information Officers (CIOs). CIOs are better equipped to address AI ethics practically, considering factors such as risk tolerance, regulatory compliance, and potential impacts on business operations. According to Accenture, only a small percentage of organizations have successfully assessed risks and implemented best practices on a large scale.

Key real-world issues surrounding AI governance include lack of transparency, bias problems, accuracy issues, and challenges with defining the boundaries of AI purposes, according to Gartner analyst Frank Buytendijk. Regulations like GDPR dictate that data collected for one purpose cannot be repurposed using AI for another purpose without proper consent.

Buytendijk highlights an example: “If you are an insurance company, it is problematic to use social media data to identify smokers through AI analysis, especially if the applicants stated otherwise in their insurance application.”

Despite the looming threat of AI-specific regulations, ensuring alignment of AI tools with an organization’s core values and business objectives goes beyond mere compliance. Forrester principal analyst Brandon Purcell emphasizes the business benefits of ethical AI practices, stating that aligning AI objectives with real-world outcomes leads to increased profitability, revenue, and efficiencies.

Paula Goldman, Chief Ethical and Humane Use Officer at Salesforce, agrees, stating that building trust into AI systems enhances their effectiveness and productivity. She emphasizes the importance of AI ethics in ensuring optimal performance and productivity.