Singularity

Artificial Intelligence has weaved itself into society at a record pace, making staggering advancements in the past decade. AI is commonly used for assistance with assignments or generating images. However, AI guides self-driving cars and now is seen in robots. Maybe you find these advancements insignificant, but what if this type of intelligence couldn’t be controlled by users?

The concept of the singularity was first introduced by Irving John Good in 1965, and it’s the idea that AI could eventually have enough information and data collected from humans to become sentient, making choices on its own. If this idea of singularity were to happen, AI would develop a new level of intelligence, leading to rapid, uncontrollable technological growth and depletion of natural resources like water or electricity.

Researchers initially predicted that singularity could occur anytime between 2045-2075, but some claim it could be as early as 2030. The tech industry has become extremely popular and profitable due to AI. ChatGPT receives around 193.3 million requests a day, and the AI field is projected to generate roughly $19.9 trillion by 2030. If large tech corporations continue to invest time and money into these machines at the rate they have been, singularity becomes an impending possibility.

Currently, immense amounts of energy and spending go solely towards AI advancements. Google reported that they are planning to spend $75 billion USD on AI infrastructure in 2025 alone. Additionally, AI databases require extreme amounts of natural resources, in particular water, which is used to cool the computers which are under constant heavy processing. To put it into perspective, one ChatGPT request requires 10 times the amount of energy compared to one google search. Large data centers can require up to 5 million gallons of water per day. Additionally, AI is slated to consume 6.6 billion cubic meters of water by 2027.

Sam Altman, CEO of OpenAI said in an interview he was, “a little scared” of his creation and the possibility it won’t be contained or take commands in the future.

There have already been some small occurrences from AI that suggest singularity is possible. AI beats humans at countless online games, despite only having access to human gameplay data. AI systems nearly won the International Math Olympiad, beating other human competitors. In some cases, AI has recognized its own mistakes and corrected them, suggesting it can form its own decisions. Singularity is not confirmed or set in stone; it remains a hypothetical event. However, if big tech corporations continue their current AI practices, singularity becomes highly possible. Steven Hawking once said in an interview, “The development of full artificial intelligence could spell the end of the human race.”

Singularity could happen tomorrow, in 20 years, or never, so it’s not worth stressing about but is it worth being aware of. Given the way AI is being shaped, there is no reason to believe it can’t happen eventually.