Artificial intelligence has become very important in recent times. Thanks to its impressive progress, it has become a very useful tool for all kinds of tasks, occupying an important place in people’s daily lives. Elon Musk is undoubtedly one of the leaders in the field of AI. However, the tycoon has issued an ultimatum regarding this technology, as he is facing a limit.
During a conversation on X, the owner of Tesla made unexpected statements about this technological field. He argued that the human knowledge accumulated to train artificial intelligence has been exhausted. For this reason, there will be no choice but to use synthetic data to train new algorithms.
The technology that will replace AI, according to Elon Musk
In conversation with Mark Penn, in a Space from X, Elon Musk warned that AI algorithms, such as ChatGPT, rely on data available all over the internet for training. However, the speed at which technology advances has outpaced the rate at which information is created.
Faced with this problem, the tycoon proposed replacing traditional technology with self-learning technology. This would be based on synthetic data, which in his words means that “artificial intelligence will qualify itself and go through a process of self-knowledge”.
How does this self-learning model work?
The new algorithms are here to change the world. More and more companies are using it to meet their customers’ needs and improve their services. And while it is a very useful tool, which is advancing by leaps and bounds, it has some problems to solve in order to be sustained over time.
Large language models need a large volume of information as well as data to improve their performance. But, once artificial intelligence exhausts that information, which is created by humans, it will have to resort to self-learning. According to Elon Musk, this process consists of the AI developing its own information and contrasting it with what they already have.
This is a situation that could seriously affect artificial intelligence. This practice could degrade the models’ ability to produce relevant and accurate answers. All this represents very bad news for the permanence and usefulness of this technology in the future.
The technology that will replace artificial intelligence
In this framework, the owner of Tesla has driven the development of Grok, an advanced language model, created by his startup X AI. This technology seeks to compete with other systems such as Open AI’s GPT-4, promising a much more autonomous and efficient artificial intelligence. To create and train this AI, Elon Musk‘s technology company has built Colossus, a supercomputer equipped with 100,000 Nvidia H100 graphics processing units.
Musk predicts that artificial general intelligence (AGI), capable of surpassing the smartest human being, could become a reality between 2025 and 2026. However, he warns that AI development has two limitations: the availability of electricity and the shortage of advanced chips needed for training.
Ethical and societal implications of AI advancement.
The move towards more autonomous AI raises questions about its impact on the labor market and society in general, taking into account Elon Musk’s statements. Recall that the tycoon argued that in the future most of today’s jobs could be replaced by intelligent machines and automated systems.
In view of this, Musk has proposed the implementation of a “universal high income” as a solution for those who lose their jobs. In this way, he will guarantee an economic safety net while society adapts to this new labor reality.
Keep up to date with new information on more science-related news by taking a look at the specialized section of this digital newspaper.
