What Happens If AI Goes Out of Control?

Artificial intelligence has gained popularity in recent years. The tools created and used with this technology enable more to be accomplished with less money, time, and energy, freeing up resources for more productive uses.  While the technology deserves all the appreciation as it clearly comes in handy, there are still concerns that an ordinary user may not be aware of. For example, what if the system that was deliberately developed to learn from all of its encounters ends up mastering enough competence to surpass human intellect and spirals out of control? Will mankind be able to keep it under control? Or will AI eventually take over the control of humans?  Let us see what this MIT Technology Review article has to suggest regarding the potential risk of “If AI Goes Out of Control” and whether we can avert it. 

According to the article, while even the biggest tech corporations want to use AI technologies like generative AI, it will most likely just be a matter of time until things go wrong. The article suggests that AI models can still be manipulated to violate their own rules. They are nevertheless subject to assault, and there is little to prevent them from being exploited as misinformation, fraud, and spam tools. Companies must recognize that they have an ethical, moral, and legal responsibility to guarantee that their products are safe for customers to use. As a result, adequate control over AI use is still essential. The article also highlights that members of the European Parliament recently approved draft legislation calling for a ban on face recognition technology in public spaces. It also prohibits predictive policing, emotion recognition, and arbitrary internet scraping of biometric data. As a result, it is probable that the EU will enact further restrictions to regulate generative AI, and the parliament wants corporations developing massive AI models to be more open. However, according to the article, the EU is still a long way from enacting generative AI laws, and many of the suggested parts of the AI Act will not make it into the final form. Finally, the article implies that the historical damages caused by such technologies are expected to increase exponentially as generative AI becomes more interwoven into our society. If ever AI goes out of control, then we are still unsure of how much damage to expect. 

Imagining a world in which AI goes out of control is terrifying in itself. The previous text highlighted potential hazards and asked if we were prepared to confront them.