Behind the Algorithm: How AI Fuels Exploitation, Conflict, and Crisis

You probably heard people saying how AI will take over the world and wondered how that is possible. The truth is AI is not always a nice and friendly tool; sometimes those patterns of information can be destructive and harmful, like blackmailing its users. 

Since AI is a neutral tool, it does not only intake good patterns of human behavior but also bad ones. And unfortunately AI companies have not been taking a responsible role in ensuring human rights are being respected. Here is why. 

• Destructive AI Behavior: Claude Opus 4, a highly advanced online AI tool, in particular, has been cited by the news for blackmailing its engineers who threaten to remove it. During a test of the Claude Opus 4, there were fake prompts and information given about an engineer's marital affair and that the AI system will soon be removed and replaced, leading to the blackmail behaviors. 

• AI sweatshops: If AI sounds so very human, it is because millions of human data have been inputted to allow it to understand these human patterns. Unfortunately, this work is tedious and intensive, often outsourced to third-world countries for very little money. Data laborers in South America and East Asia were reported to work 18 hours a day for wages below the countries' minimum. 

• Work Trauma: Many international AI data sorters must filter out harmful graphic content into learning patterns, which is outsourced unethically. ChatGPT is reported to use young workers from Kenya for less than $2 per day to sort out violence, hate speech, and sexual abuse, reporting deep trauma to these workers. 

• Fueling Water Scarcity: Tech Giants Amazon, Microsoft, and Google are using vast amounts of water from parts of the world facing water scarcity to cool data centers. 42% of water used by Microsoft is by places with high water stress. Amazon’s three planned data centers in Spain’s Aragon region have been approved to draw about 755,720 cubic meters of water annually, an amount comparable to what would be needed to irrigate 576 acres of corn, one of the area’s key agricultural crops. 

AI may be a helpful tool, but it is definitely not a dainty daisy when it comes to ethics, and giant tech AI companies profit off of exploitative labor practices and questionable violations that are not just towards their laborers but also victims of AI weaponization. 

AI Weaponry and Destruction in the Middle East

Globally there is so much conflict in Middle Eastern and Muslim countries it seems coincidental, as if Trump is fueling this conflict on purpose. The reality isn't too far from that, with not only US weapon companies like Lockheed Martin fueling genocide but also AI tools being used as weaponry through destructive AI projects.

• Project Maven: A US military AI tool project that was used by the Pentagon to identify targets for more than 85 air strikes in the Middle East in the month of February 2024. This system is being used for places like Yemen, Syria, and Iraq. But the Pentagon assures us humans are checking the AI recommendations. This has revolted thousands of Google workers, leading to the company not renewing its contract. 

• Project Nimbus: Google and Amazon have partnered in a billion-dollar contract with Israel, a country recognized by the UN for its ongoing persecution, genocide, and violation of human rights towards Palestinians. This project will use AI technologies like surveillance, facial detection, and object tracking as a means of aiding Israel with its affairs, and the companies are not allowed to cancel the contract due to public pressure or prevent arms by the government. 

Drone technology is powerful and used vastly by governments, accounting for a great portion of modern deaths on the battlefield. With the help of AI, weapons of destruction can strategically destroy more innocent civilians. 

If you ever wondered why AI technology is a double-edged sword technology tool, here is why. The world needs responsible companies controlling AI more than ever before.