Something remarkable occurred on the U.S. stock market on June 5th – Nvidia reached a $3 trillion valuation, primarily due to the success of its AI-focused semiconductor chips. This milestone previously belonged to software giants like Apple and Microsoft. However, Nvidia’s exceptional profit margins from AI chips propelled it into the $3 trillion club.
Despite this, cautionary advice was sounded by ARK Investment head Cathie Wood. She highlighted the necessity for AI to demonstrate its value in various domains for Nvidia to sustain its valuation. The future success of AI companies such as OpenAI, Microsoft, and Alphabet in monetizing AI remains uncertain, with questions surrounding profitability in the realm of AI services.
Taking ChatGPT for a test drive
Curious about ChatGPT’s capabilities, the author decided to explore the technology for automated tasks. Collaborating with a computer science graduate to develop an application interface, they ventured into ChatGPT’s pricing structure.
OpenAI, the owner of ChatGPT, offers its output in “tokens.” These tokens are priced between $0.02 to $15 per million tokens, depending on the required model. Opting for the GPT-3.5 Turbo model at $1.50 per million tokens seemed ideal for the task at hand.
Delving deeper, the author inquired about ChatGPT’s energy costs. Despite a convoluted response, it was estimated that answering a query cost approximately $0.00078, a minute fraction of the token cost. This leads to the consideration of OpenAI’s expenses in providing AI services, which comprise not only energy costs but also training models and procuring AI chips, potentially from Nvidia.
While upfront costs are significant, the scale of operations could enable companies like OpenAI, Microsoft, and Alphabet to profit, especially with evolving competition in the AI chip market. The optimization of costs, coupled with increasing demand for AI services, paints a promising picture for these enterprises in achieving profitability over time.
Thus, the prospect of AI companies monetizing their ventures is not far-fetched. In the grand scheme, this development not only benefits the companies themselves but also bodes well for supporting industries such as Nvidia, hinting at a mutually advantageous ecosystem in the AI realm.
Suzanne Frey, an executive at Alphabet, serves on The Motley Fool’s board of directors. Rich Smith holds no positions in the mentioned stocks. The Motley Fool has vested interests in and recommends Advanced Micro Devices, Alphabet, Apple, Microsoft, and Nvidia. It also suggests Intel and certain options related to these companies. The Motley Fool upholds a disclosure policy.
