Nvidia founder and CEO’s jacket sold for a fabulous price. Lithuania, a country that first got rich selling used rags (unicorn No. 1), has soared to the skies. Well, what is AI – selling used junk. Everything else is a scam – poor people from Africa and India answer people’s questions for pennies.
Nvidia CEO Jensen Huang’s leather jacket was sold at Sotheby’s for an impressive $960,000, with all proceeds going to the Edge Institute, a non-profit organization supporting young technology developers. While our comparison to Lithuania’s first unicorn Vinted and the second-hand clothing culture is extremely witty and apt, the real business of artificial intelligence (AI) is not in textiles, but in hardware and global infrastructure.
What is important to know about this event and the reality of AI?
• The value of the jacket: A Tom Ford jacket was sold for around $10,000, with an initial estimate of up to $60,000 at auction. Collectors have raised the price because of its historical significance to the AI boom.
• Nvidia’s power: This company is not in the clothing business. It makes the world’s most powerful graphics processing units (GPUs) (such as the H100 or the new Blackwell), which are essential for training and running AI models (such as ChatGPT). It is precisely because of this hardware that Nvidia’s market value has exceeded $4 trillion.
• Data annotation (human labor): Our observation of workers in Africa and India has a real basis, but it is not a “hoax” but is technologically called data labeling and reinforcement learning (RLHL). Huge teams around the world manually check, correct, and sort information so that AI can understand context and avoid mistakes. This is the foundation on which algorithms are built.
The AI industry consists of three main parts, which explain where the money really goes:
Layer What does it consist of? Real-world example
Hardware Semiconductors and supercomputers: Nvidia chips that made the CEO’s jacket iconic.
Data and Work: Huge databases and their filtering. The thousands of specialists in developing countries who tag text and images for pennies.
Software Algorithms and end-user applications: Large language models (LLMs) that generate text, code, or images.
Komentarų nėra:
Rašyti komentarą