2023 was the year of artificial intelligence (AI). ChatGPT unexpectedly becomes the fastest growing app in history software and catapulted investments to one point technology It is as hopeful as it is controversial. The fire created by Silicon Valley giants grew at a dizzying pace and spread to the markets, and the sector experienced its best year since the dotcom bubble.

But 2024 points in a less rosy direction. Analysts predict that the next 12 months will be a “cold shower” for the booming manufacturing industry. Generative AI. “The excitement is so great that we believe it is overrated and many hurdles will have to be overcome to bring this to market,” said Ben Wood, principal analyst at market analysis firm CCS Insight.

These are the three big challenges facing the technology market. artificial intelligence this year.

material shortage

Predictions about a cooling in the industry are based on a very important issue: Their costs are too high. Artificial intelligence does not work without it chipsComplex electronic circuits that act like your brain. The emergence of productive technology has triggered demand semiconductors by large companies such as Google anyone Microsoftto foster a global struggle to lead this market.

However, the dizzying increase in orders also creates problems for the few companies that can produce chips. Taiwanese giant TSMCA world leader in its sector, famine The degradation of its packaging method will prevent it from meeting all customer demands for the next year and a half. A reality that stands in contrast to the increasing adoption of artificial intelligence in business by 2024.

It also requires semiconductors. critical minerals And rare earthChemical elements that are complex to extract and may in some cases face shortage problems. Its strategic use and growing demand are increasing geopolitical tensions, especially tensions between them. United States of America And Chinese– threatening to cause disruptions supply chain spherical.

climate impact

Although it seems like a spiritual being, artificial intelligence requires processes that consume enormous amounts of energy. energybut also This. As the popularity and demand of this technology increases, climate impact From the infrastructures that support this. H100 chips produced by the American company NVIDIA One of the most powerful graphics processors in the world, which plans to sell in 2024, these processors will be able to consume the same energy as small countries such as Lithuania or Guatemala within the year. The fact that this trend will continue to increase over the next 12 months worries many experts.

political manipulation

2024 will be a critical year for Türkiye democracy. Nearly a billion people (more than half the world’s population) will participate in the coming months. elections It will be held in 76 countries, including the USA, India, Russia, Mexico, England, Indonesia and the USA. European Union (EU). Artificial intelligence will probably play a role in all of this.

The commercial proliferation of increasingly powerful generative models has made it easier and cheaper than ever to create realistic content, from images to sound. This democratization of access to artificial intelligence political propaganda, a communicative shift that parties of all kinds have already experienced. For example, American Republicans released a video imagining apocalyptic scenarios and linking them to the re-election of Joe Biden.

However, its use also brings risks as it accelerates the spread of awareness campaigns. Lies It aims to manipulate public opinion. This aim can be achieved by producing synthetic images or ‘deepfakes’ that will specifically distort reality and harm the political opponent. Additionally, errors produced by some of these AI tools (“hallucinations‘—can also help strengthen disinformationeven if unintentionally.