The End of Authenticity

How the Increasing Utility and Complexity of AI will Shape our World

B. N. Dent

Ukraine is currently facing a number of challenges and political tensions. The country is still in conflict with Russian-backed separatists in the eastern regions of Donetsk and Luhansk, despite several ceasefire agreements. The economy has struggled, with high inflation and unemployment rates, and corruption remains a widespread problem. In addition, the government is facing pressure from opposition groups, civil society, and international organizations to address these issues and implement reforms. Despite these challenges, Ukraine is making progress in some areas, including its efforts to integrate with the European Union, and its people continue to exhibit a strong spirit and resilience in the face of adversity.

I am sure you are confused as the preceding paragraph has no correlation to what you may have expected this article to be about based on the title and subheading. The paragraph above is merely an example of the immense strength and level of intricacy that computer generated responses currently have. Specifically, the paragraph was generated when ChatGPT was asked to “write a paragraph about the current situation in Ukraine.” It is worth noting that the response was solely reliant on information prior to 2021 due to that being the most recent time that the data library accessible by ChatGPT was updated. For those who haven’t already heard of ChatGPT, it is essentially an online chatbot designed to generate human-like responses to any question that may prompt it. Questions can range anywhere from extremely complex mathematical problems to requests for generating a unique creative story. 

So how does this insane technology work? ChatGPT functions via transformer architecture whereby when generating a response, it uses the context of the input (question used to initiate the response) and the previous word or sequence of words it has already produced in order to calculate the most likely word to follow the sequence. In order to perform the calculation of finding the most probable word to generate in the response, the bot utilises its access to an extremely large amount of text data which allows it to understand which word is most likely to add continuity to both the coherence of the sentence structure and to the relevance of the response. The predominant and unmatched benefit that ChatGPT provides at the moment is the specificity and accuracy that it is capable of providing on a large range of subject matters. We have all been in a situation where we have a quite specific query, so we research online to seek a response, but whilst we may find something that vaguely resembles what we were hoping to find, we do not find a definitive response. It is here, with great precision and accuracy, where ChatGPT thrives. On the flipside, some of the disadvantages include its inability to offer unique insights, lack of understanding regarding how the world works which can lead to nonsensical responses, and limited context due to the finite amount of data in its reach and the frequency at which its accessible data is updated.

In the past, we have seen that the emergence of powerful technologies has had the ability to completely change the world that we live in by advancing our technological state and shaping how we interact within society. The societal implications of easily accessible advanced AI whilst seemingly beneficial may prove to be harmful due to the dependence that can be formed with such technologies. The dependence on AI to think critically, research, explain and come to conclusions for us is an extremely dangerous threat, ultimately leaving us without the ability to think for ourselves. The pollution of this innate human characteristic can lead to a repression in creativity and genuine authenticity. Alongside AI oriented apps like ChatGPT are others such as DALL-E which is able to generate “unique” images from text. Whilst the by-products of these apps such as stories and artworks are claimed as new ideas, they are all based off of pre-existing ideas stored in a database, in effect creations of inauthentic authenticity.

“The pace of progress in artificial intelligence is incredibly fast. Unless you have direct exposure to groups like Deepmind, you have no idea how fast—it is growing at a pace close to exponential. The risk of something seriously dangerous happening is in the five-year time frame. 10 years at most.”  —Elon Musk wrote in a comment on Edge.org