One of the most pressing concerns related to science and morality in the contemporary world is the growing capabilities of artificial intelligence that could potentially threaten the whole human race. These models learn from the world around them and might eventually become sentient, and it is far from certain that they will decide to be benevolent to humans. In my personal opinion, it is crucial to develop additional safety and precaution regulations to prevent AI from hurting people.
There is extensive evidence that demonstrates the unlimited capabilities of artificial intelligence. This threat occurs due to a concept of “mimicked evolution,” meaning that deep learning models learn and evolve similarly to humans, absorbing all information around them (Marr, 2020, para. 1). Even though scientists program AI to be benevolent and never hurt humans, it is impossible to completely prevent artificial intelligence frameworks from getting dangerous information. After all, humans themselves are not perfect, while violence and prejudice are still present in contemporary society. From my personal observations, if deep learning algorithms are not restricted in their education mechanisms, they might learn the negative aspects of today’s society. Moreover, in her novel, Mary Shelley demonstrated this threat in the example of Frankenstein’s monster, which was created to help humans (Shelley, 1818). However, the creature was corrupted by the world around him, eventually becoming hostile and dangerous. This argument clearly shows how sentient entities, whether AI or Frankenstein’s monster, might become violent and threaten people around them.
The scientific controversy between AI risks and benefits is a highly relevant topic. The example of Frankenstein shows that even more than 200 years ago people were concerned about mimicked evolution and the potential dangers of artificially created sentient beings. However, unlike Frankenstein’s monster, it is possible to implement more safety regulations concerning contemporary AI models. Therefore, humanity must consider the risks and benefits of AI development and ensure that these frameworks will never harm people.
References
Morr, B. (2020). Is artificial intelligence (AI) a threat to humans?Forbes. Web.
Shelley, M. (1818). Frankenstein. Lackington, Hughes, Harding, Mavor & Jones.