The AI Paradox

The AI Paradox

James Gorrie

If the rush to adopt artificial intelligence (AI) seems familiar, it’s because we’ve seen this movie before. In the late nineties, organizations of all sizes, but especially large or enterprise-level businesses, struggled to make the transition from legacy functionality to digital transformation. As late as the COVID-19 pandemic and global lockdown in 2020, enterprise organizations were rushing to complete their digital transitions, particularly when it came to enabling and protecting widely distributed remote workers with home office operations, workflows, and compliance regulations.

What’s more, an AI arms race is well underway, as Microsoft, Apple, and other tech firms invest billions into AI products such as OpenAI. Everyone knows full well, or should, just how intense the race to dominate AI is at the global and corporate levels. It’s that transformative.

AI Altering Business Practices

There’s no question that AI is the next technological advancement that will fundamentally alter our way of life in the very near future, and perhaps for evermore. In fact, it already is and continues to do so. But in these early and heady AI days, there are practicality speed bumps, too.

For example, corporate executives and boards are encountering implementation and control roadblocks that are preventing their organizations from realizing more than just a fraction of AI potential, not to mention justifying the millions of dollars in spending that an enterprise-level AI deployment costs in streamlining their operations, optimizing workflow, harmonizing siloed divisions, and other organizational challenges that enterprise-level organizations struggle to handle well.

But that’s just the beginning of the challenges and risks that AI poses to even mid-size businesses as well as enterprise-level organizations.

The Risk to Privacy and Ethics

The promise of AI lies not only in its speed of data analytics but in its utter power to quickly access—and potentially abuse access—to personal information. Individuals’ private data are shielded by their legal right to privacy, and the responsibility to protect that privacy falls upon organizations that possess private data. If an AI-driven program or product is abused or even just not monitored, the risk of an enterprise finding itself in violation of privacy laws rises.

There’s also the risk of bias perpetuation and the resulting social impact of any biases that are baked into the AI program. These may well include the personal beliefs of the AI programmer or servicer, with outcomes that harm certain people with personal beliefs that are counter to those that are in the AI programming.

Unlawful surveillance is another risk that AI brings to organizations of all sizes. Do people have the right not to be surveilled when engaging in activities on their time? Do they have the right to their behavior not being commoditized with predictive analytics and sold to the marketplace repeatedly? They should, but in practice, it’s much more a case of technology outrunning our legal system’s ability to address the new challenges and risks. The risk of technological change and capabilities is just too fast.

More...