ChatGPT is helping millions of people in different ways. Many people use it as a professional tool to help them ease their workload. However, others are effectively using it for 2 to 3 jobs at a time. Whether you are a content writer, a programmer, a coder, or belong to any profession, ChatGPT can help you.
Developed by OpenAI, the AI-based tool is a neural network. It uses a massive amount of textual data. The data comes from various textbooks, articles, web pages, and Wikipedia. Though ChatGPT is an AI-based tool, it can adopt a human-like conversational tone to generate and answer their queries. Apart from various simple queries or commands, it can engage itself in answering complex and creative text.
Unveiling the Enormous Operational Costs Of ChatGPT
We only pay our fixed amount of $20 per month. But at the back, it costs OpenAI millions to operate ChatGPT each month. Yes, that’s right, an analyst told “The Information,” confirming that it is becoming very costly for OpenAI to operate its daily and monthly servers.
Dylan Patel, at a semiconductors research firm known as SemiAnalysis, told “The Information” in a statement that OpenAI is spending loads of money to run their servers at ChatGPT. He estimated the amount to be $700,000/per day and $21 Million each month. OpenAI is spending this amount to fulfill the computing power to calculate and run its user’s prompts.
Patel further told the tech publication, “Most of this amount runs the gigantic servers the ChatGPT is using to run itself.”
Balancing Subscriptions And Server Expenses: OpenAI’s Financial Challenge
In a call with the Insider, he further said that this cost may have increased now because the estimated $21 million per month operational cost was about the GPT-3 model. But, now that GPT-4 model is already released, this cost must have increased significantly.
The discrepancy in cost estimates highlights the complexity of calculating operational expenses for AI systems like ChatGPT. As OpenAI strives to balance affordability for users and the technology’s development, maintaining a sustainable financial model remains a critical challenge. Moreover, other sources claim it takes $100,000 for OpenAI daily and $3 million monthly to keep the chatbot up and running for its users. Nevertheless, this amount might be more manageable than the users paying OpenAI as subscription charges.
Microsoft’s Role In Reducing OpenAI’s Computational Power Costs
OpenAI currently uses Nvidia GPUs for its computing power for ChatGPT and other brands. A report estimates a further need for 30,000 more Nvidia GPUs to maintain its current commercial course this year. However, Microsoft, currently one of their main collaborators and investors, is working on developing cheaper AI chips to reduce OpenAI’s cost burden.
Microsoft calls these cheaper AI chips Athena, as reported by “The Information.” According to their $1 Billion deal a year, Microsoft requires OpenAI to run its server models on their Azure cloud servers.
Today, after four years, more than 300 Microsoft employees are still working on their new AI chip. The Information again reported after confirmation from two sources stated that “The new AI chip will release for both their internal use by the earliest of next year in 2024”.
