This is not investment advice. The author has no position in any of the stocks mentioned. Wccftech.com has a disclosure and ethics policy.

NVIDIA shares hit their local nadir back in October 2022, spurred by persistent weakness in gaming and data center segments. However, the recent hype around OpenAI's ChatGPT and generative AI has flipped that bearish calculus on its head, spurring even professed bears to throw in the proverbial towel.

As a refresher, NVIDIA has jumped onto the "large language model" bandwagon via its AI-as-a-service business model. Under this model, NVIDIA will soon allow cloud-based access to its AI-training supercomputer. Further details are expected to materialize at NVIDIA's upcoming spring GTC in March.

For those who might be unaware, OpenAI's ChatGPT and Google's Bard leverage deep learning via Generative Adversarial Networks (GANs) to create new content that is virtually indistinguishable from human-created content. To do so, GANs use two types of neural networks, where a generator creates new content while a discriminator evaluates that content and then provides feedback to the generator. In this way, GANs progressively self-evolve.

MORGAN STANLEY: “While our views that $NVDA would see tactical numbers challenged in both gaming and data center have largely played out, the development of generational #AI is too much of a megatrend to get distracted by tactical concerns. The stock will .. be hard to ignore ..” pic.twitter.com/66rUXJZb8H

— Carl Quintanilla (@carlquintanilla) March 17, 2023

This brings us to today when Morgan Stanley has upgraded NVIDIA stock to an Overweight rating, pegging a $304 share price target. The new price target constitutes an increase of 19 percent relative to Morgan Stanley's previous stock price target of $255. NVIDIA is currently trading at around $260 per share in pre-market trading.

While explaining the rationale behind today's upgrade, Morgan Stanley analyst Joseph Moore noted that headwinds for NVIDIA around gaming and data center have largely played out but are now being overshadowed by the "development of generational (or is it generative?) AI."

As per a recent commentary from KeyBanc, the ChatGPT generative AI model is itself using around 10,000 A100 GPUs, contributing around $100 million directly to NVIDIA's top-line metric. Bear in mind that each A100 GPU delivers up to 5 Petaflops of AI performance.

Moreover, as we noted in a recent post, the number of GPUs that will be needed to power the ongoing commercialization of OpenAI's ChatGPT is expected to reach above 30,000.

Coming back, with NVIDIA's AI-as-a-service business model soon to become a reality, Moore noted:

"… with NVIDIA having dominance in the training market that is likely to persist for several years even with a relatively fixed number of model developers, model complexity plus multiple languages should still drive 3-5x growth in training over 5 years."

As a punchline, Moore stated that the NVIDIA stock "will continue to be hard to ignore in an otherwise challenging semiconductor environment."