admin
TECHNOLOGY: What recent scientific discovery or technological advancement do you believe will have the most significant impact on society in the next decade, and why?
Without a shadow of a doubt, its Artificial Intelligence (AI). While AI is mainly known at the moment for faking student homework, doing deep fake videos, creating pictures of cats and allowing lazy people to post answers on Quora without understanding what they are posting, it has enormous practical, useful applications.
AI is not powerful enough to solve climate change, overcome systemic prejudice and bring about world peace, but, as a tool, it enables us to manage the vast amounts of data, which we collect but don’t utilise.
Consider: at the moment, your smartwatch can monitor your vital signs.
An expert system (of the kind they started developing in the 1970s) can take that data and warn you if you are at risk of atrial
fibrillation, or even alert you that your resting pulse rate has developed a new trend. But AI could gather the data from a million smartwatches, connect it with patient case histories and use the information to develop new models, which predict things that no one has ever thought of before.
We are fascinated with AI as consumers because of its ability to create patterns based on previous patterns. However, we all know that these patterns are often imperfect and sometimes laughably, so AI may never reach the point where we can rely on the patterns it generates in critical situations. But AI’s ability to spot patterns is far more interesting and useful.
Just 24 months ago, AI-produced illustrations were rudimentary and unconvincing. Twelve months ago we were used to spotting extra fingers on AI-generated images of humans. It took a lot of user-knowledge in “prompt engineering” to get reasonable results. Now, AI-generated images are only limited by the knowledge of the person selecting them—for example, the laughable “historical” images posted by alleged “historical” spaces on Facebook are wrong not because the AI couldn’t make them right, but because the person who requested them and posted them didn’t know enough about history to know they were impossible.
This is not going to change: the speed of AI advancement is going to be faster and faster, but it will still require user knowledge. That’s why its primary impact will be felt in medical technology, product engineering, data analysis and coding—areas where the output can be tested, and the people operating the tech are themselves experts. In those areas, it will change our world. (Source: www.quora.com)
Online Reporter