A question that becomes even more curious when paired alongside the current incentives driving the market, which ultimately affect how AI is built and expanded.<\/p>\n
To understand how we got here, let’s first explore a little-known concept called “planned obsolescence.” A popular market term that describes the process of building products with a limited lifespan. Popularized in the 1920’s by General Motors through annual car redesigns, it is a practice that over the decades has trickled down across all industries, it’s why modern refrigerators only last around twelve years on average, unlike my grandmother’s fridge from the fifties that ran well into the early 2000’s, which seems unfathomable when you really think about average product lifespan nowadays.<\/p>\n
Another more recent example is a 2025 lawsuit where the FTC and the Attorney General of Illinois and Minnesota sued John Deere because farmers who spent hundreds of thousands of dollars on machines, were unable to use these machines the following year because it required a special software update, otherwise the machine was deemed unusable without this additional purchase.<\/p>\n
Which leads me to my first point, tech in its current state is not built to last.<\/p>\n
What I think most people outside of tech do not realize is that software is not something you build once and leave alone. It has to be constantly monitored because it is perpetually dealing with regressions, degradation and updates, such as unannounced software changes pushed into production by companies that lack the common decency to tell anyone what had changed, schema drift, and good old-fashioned pipeline failure, and all the other small indignities engineers spend their days cleaning up.<\/p>\n
So while tech leaders are painting doom and gloom, those of us in the muck of it are left scratching our heads with, how? Since these so-called murder bots are one update away from failure, which can be caused by a simple typo in a script buried on the backend of some engineer’s repo, it’s a thing that happens constantly, and since planned obsolescence isn’t going anywhere, the same incentives driving markets, ultimately drive AI.<\/p>\n
Leading to my second point, infrastructure, more specifically, the much hated data centers. Since AI isn’t some nebulous intelligence floating in cyberspace, it requires land, electricity, hardware, labor and an exorbitant amount of capital driven by the same incentives, which is leading to all the problems we are currently facing. Such as how data centers are using an excessive amount of fresh water, which I’m sure many may not know is not a necessity, but a choice. A choice driven solely by KPIs and the need to get things done as quickly and as cheap as possible. Which has left corporations to default to primitive ancient techniques used by humans for thousands of years.One of the earliest documented examples comes from Ancient Egypt, where people used fresh water stored in porous clay vessels to draw heat from the surrounding air. For thousands of years, humans have turned to water as a means of cooling: Ancient Persians built vast underground channels that carried groundwater to windcatchers, helping to pull cooler air up into rooms, while the Romans relied on substantial quantities of fresh water across water-intensive trades, from pottery production to metalworking, where dedicated aqueducts helped cool hearths and smelting furnaces.<\/p>\n
Which leads me back to my original question of why are we, after thousands of years of engineering advancements, fallen back onto old cooling techniques, especially in places where freshwater in itself is already scarce. While yes, the reason is market incentives and the need to build as cheaply and as fast as possible, I do want to point out that there are alternative solutions.<\/p>\n
One example being immersion cooling; which is the use of a nonconductive dielectric liquid to cool down servers. A practice that has been around for decades, and involves submerging entire infrastructure into mineral oils or some non-conductive dielectric fluid.<\/p>\n
Microsoft ran a study back in 2020 where they submerged servers into mineral oil and found that the machines could operate normally while receiving better cooling than under conventional air-cooling systems. They again repeated the study in 2025 using a special dielectric fluid, which showed a wealth of savings, not only in the cooling process, but also in cutting energy use by 15–20% and greenhouse-gas emissions by 15–21%.<\/p>\n
Showing that AI doesn’t inherently require freshwater cooling. No, it’s humans making these decisions based on a series of driving market factors.<\/p>\n
Leading me to my third and final point, of how AI is being paraded around as this autonomous entity. Like most things in life, it takes a lot of effort to get something to work, whether it’s building machines or getting a script to run, it takes labor, specifically human labor. So, when I hear stories about how an AI jumped its sandbox and found its way onto the internet. I find myself yet again questioning how, because scripts aren’t cognitive, it still requires rules, meaning for AI to have ended up on the internet, someone had to program it to do such. Which leaves me feeling like I’m watching a scene out of the Wizard of Oz, where a man is sitting behind a curtain speaking through a mechanical mask. This I feel is the current theatrics currently dominating media around AI. Another grating example is an AI researcher at Anthropic recently resigned and cited his reason for leaving were due to the race toward “self-improving superintelligence [that] could end humanity by the end of the decade”, which on the surface sounds daunting until you dig into the actual paper Anthropic recently released called “Detecting and countering misuse of AI: September 2026”.<\/p>\n
From the title itself, one can already deduce the overall theme of the paper, which pretty much lists out all the recent bad actors who used artificial intelligence to do malicious things, ranging from state-linked groups and commercial spyware vendors to criminals and politically motivated individuals using AI for espionage, fraud, propaganda, surveillance, weapons development, etc, which isn’t anything new. There’s always been bad people who seek to do awful things with whatever latest tool is at hand. However, while intent hasn’t changed, the paper stresses that what usually took a lot of labor, expertise, time and money can now be done by a single individual.<\/p>\n
But that is not how most articles are being framed. Instead, it’s leading with the idea that a rogue machine came up with a plan and sought out to execute that plan, which is simply not true. A person came up with an idea and used a machine to carry-out that idea.<\/p>\n
Which is why this current narrative feels like we’re assigning intention to machines rather than examining the people and goals behind them. And while AI systems can sometimes behave in sporadic and unpredictable ways because of bad code, flawed design, or unintended interactions, that doesn’t inherently suggest cognitive intent. I think, collectively, moving forward, our questions should first identify the incentives, then ask who is building these systems and what problems are they trying to solve. Only then will we be able to identify whether AI is really trying to kill us, or if it’s just people being people.<\/p>\n
About Aisha Yesmine<\/p>\n
Aisha Yesmine is associated with commentary on artificial intelligence, technology infrastructure and the broader societal implications of emerging technologies. Her analysis focuses on the relationship between technological systems, commercial incentives and human responsibility.<\/p>\n
Media Contact<\/span> UNITED STATES — October 13, 2026 — Aisha Yesmine In today’s media landscape, it seems everywhere we turn there’s a story about how AI is making the planet ultimately worse—or,<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"_links":{"self":[{"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/posts\/618035"}],"collection":[{"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/comments?post=618035"}],"version-history":[{"count":0,"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/posts\/618035\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/media?parent=618035"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/categories?post=618035"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.olympiajournal.com\/news\/wp-json\/wp\/v2\/tags?post=618035"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
Company Name:<\/strong> Aisha Yesmine<\/a>
Contact Person:<\/strong> Aisha Yesmine
Email:<\/strong> Send Email<\/a>
Country:<\/strong> United States
Website:<\/strong> https:\/\/www.aishayesmine.com\/<\/a><\/p>\n<\/p>\n","protected":false},"excerpt":{"rendered":"