Harvard Business Review: “4 Types of Gen AI Risk and How to Mitigate Them”

From Harvard Business Review:

Many organizations, however, are still understandably hesitant to adopt gen AI applications, citing concerns about privacy and security threats, copyright infringement, the possibility of bias and discrimination in its outputs, and other hazards. The downsides of widespread adoption of gen AI range from mildly annoying (such as personalized spam) to truly catastrophic (such as the fast depletion of water sources from vulnerable parts of the planet to support the largest data centers ever constructed). Some organizations have banned their employees from using it. Apple and Samsung, for instance, have banned the internal use of ChatGPT, especially by the software development team, after realizing that potentially sensitive code had been uploaded to the platform, risking the loss of confidential information.

Right now, governments are scrambling to come up with reasonable frameworks and laws to manage this technology and its downsides, which means companies are mostly having to navigate these questions on their own. Clearly, all types of risks posed by gen AI are not the same, and we thus need to understand and manage them accordingly. In this article, we propose a high-level framework that will provide executives with a way of classifying the potential challenges within the gen AI landscape and then mitigating them.

A Blueprint for Generative AI Risks

We classify gen AI risks based on two factors: intent and usage. We separate accidental misapplication of gen AI from deliberate malpractices (intent). Similarly, using gen AI tools to create content is differentiated from consuming content that other parties may have created with gen AI (usage). We believe that each of the four types of risks highlighted in our framework will present distinct challenges:

• Misuse 
• Misapply
• Misrepresent
• Misadventure

Learn More, Read the Complete Article (about 2200 words)

Filed under: Associations and Organizations, Data Files, News