Artificial intelligence has been with us in some form or another for a while. It nudges us in the direction of online content, it tells us which products we might like to buy, and it works behind the scenes in just about every industry there is: from banking to farming to fashion.
The technological world has, over the last few months, been swept over by a new tide of ai-related hype. The arrival of ChatGPT, which is soon to be introduced into Microsoft Word, is just one manifestation of a broader trend. Things are going to get very automated, very quickly – and it’s worth pausing for a moment to reflect on the potential implications.
Are there any downsides towards using smart technology?
The advantages of AI are difficult to dispute. In some cases, it produces demonstrably superior results in a far shorter timeframe. It can judge which customers at a bank are more likely to repay a loan, or even generate extra frames in our favourite videogames. It can even improve quality of life for people with mobility issues, making a practical difference when it’s deployed in smart speakers and other devices.
The downsides, however, must also be weighed. AI is, for the time being, extremely costly. ChatGPT’s server costs are surely monumental. Then there is the idea that it might inhibit human creativity, presenting quick solutions to problems that would ideally require deeper contemplation.
We should also consider that machine-learning AI will inherit the biases in its training data. So, if the data is skewed in the wrong direction, we could end up with seemingly-neutral AI decisions that are racist, sexist, or otherwise discrimanatory.
That’s before we even consider the idea of an AI-powered singularity, wherein artificial intelligence rapidly improves beyond our ability to constrain it. Then we might consider the fact that ai, lacking human emotions, might make ethical decisions that are divorced from human morality. This has been a staple of science fiction for decades, from Asimov’s three-law stories to the more recent horror smash, “M3GAN.”
Overcoming these obstacles
What we do about these potential problems is a matter of some contention. Several solutions have been floated, all of which might be deployed proactively, before the worst consequences arrive.
Continuing to train human experts, and building the right data-science skillsets, is critical. Executive-level decision-makers will also need to familiarise themselves with the technology and how it works. In dealing with ethical challenges, many of which might relate to the environment, it might be a good idea for a business to collaborate with outside specialists in tech. This will help to avoid information siloing, and to ensure that decisions are made based on the best available expertise.