The world that many science fiction writers have cultivated for decades concerning artificial intelligence’s implementation and subsequent takeover is becoming increasingly realized by the day, and with that comes its impacts on the people themselves and their surroundings.
The effect of AI use on the physical environment, as well as its impact on social environments, makes its practical application both unethical and a dangerous precedent for future technology.
The energy required to power an AI data center is immense, as it involves millions of data transfers and interactions to support both the development of AI code and its generative capabilities.
According to CNBC, one AI query uses the same amount of energy as charging a smartphone. Billions of inquiries can take place at the same time at a data center. All occur using either third-party power suppliers or are generating more power out of an already established local municipal power grid, placing stress on it, according to Nuveen, a global investment company.
For example, the cryptocurrency data center in Denton, which will be converted into an AI data center later this year, will require significantly more energy to operate.
Denton, much like the rest of Texas, will begin to experience the effects of having a data center on its grid, including reduced energy efficiency and increased vulnerability to overtaxing, leading to blackouts and exacerbating climate change issues, especially during the summer.
The effects of rising Texas temperatures and summers when Texans need their electricity more than ever are unacceptable.
All of that energy usage is not just for AI capabilities but also for the maintenance of the servers and facilities that make AI generation even possible. Such generated electricity is made in a variety of ways, with companies such as Amazon, Microsoft and Google pledging to look into expanding more alternative sources of energy for powering AI, such as nuclear power.
While these promises are great, the problem is that these companies are not making the significant changes they need to make fast enough to lift the burden off the average person.
Until then, the use of natural resources is the most prevalent way to generate electricity, with natural gas being the leading resource in Texas. The use of natural gas produces fewer harmful chemicals and carbon dioxide than other sources, but it is still a significant cause of air pollution, such as smog.
Data centers also require copious amounts of water to maintain their servers. A wholesale or retail data center that is not very large in size takes up 6.57 million gallons of water a year to cool its facilities down and regulate machine temperatures.
The way that the water is used once it arrives at the data centers is not the issue. It is where the facilities are being built and how they are interacting with those communities and environments.
Clean drinking water is in high demand, and most of the world's water is being allocated for other uses, exacerbating the problem since the environment also requires water extracted from the ground.
AI builds upon this, as a fifth of the regions where data centers are built are experiencing water scarcity, such as Abilene, Texas, which is building a new data center amid standing water restrictions put in place to address its current water scarcity.
If companies seriously invested in better energy generation and water use practices as they have pledged to, AI is a tool that could be used in many ways. AI could be beneficial and promote the general ease of life if correctly implemented.
Nonetheless, the ecological strategies companies have put into place to minimize the carbon footprint and water resource use have yielded low results as the AI industry continues to grow rapidly.
Proper education about AI and changes to maintenance protocols can lead to it being a positive force in technology.
From completely automating more dangerous parts of jobs, such as construction and dangerous maintenance, to allowing more openness in the job market, AI could stand to give many opportunities in society that were thought impossible before.
Unfortunately, it is not currently possible to replace human laborers in these dangerous positions. Changes in the job market only stand to negatively impact those in need of minimum wage and entry-level positions, while saving employers more money.
Unless more legitimate opportunities in the job market open up, using AI to replace jobs in a job market that has seen the highest rise in unemployment since the end of the pandemic is incredibly irresponsible.
Individuals need to stop any heavy reliance on AI and need to demand deceleration on the expansion of AI, specifically data centers, and instead advocate for legislation that requires more conscious and ethical means of maintenance.