Nvidia (NVDA 2.29%) stock is up nearly 190% since the beginning of the year as the growth of the generative AI market sparked a buying frenzy in its high-end data-center graphics processing units (GPUs), which are used to process advanced machine learning and AI tasks.

Most of the top generative AI platforms like ChatGPT use Nvidia's GPUs, so it seems poised to profit from the expansion of the global AI market, which Grand View Research expects to grow at a compound annual growth rate (CAGR) of 37% from 2023 to 2030. Nvidia also remains the market leader in discrete GPUs for high-end PC games.

Analysts expect those growth engines to boost Nvidia's revenue and adjusted earnings per share (EPS) by 59% and 133%, respectively, in fiscal 2024 (which ends next January). Those growth rates suggest its post-pandemic slowdown -- which resulted in nearly flat revenue growth in fiscal 2023 -- is already over. But should investors still chase Nvidia as it hovers near its all-time highs? Let's compare the bear and bull cases to decide.

Nvidia CEO Jensen Huang holds a GPU.

Image source: Nvidia.

What the bears will tell you about Nvidia

The bears will warn you that Nvidia isn't the only company which produces chips for processing AI tasks. Microsoft, Alphabet's Google, and Meta Platforms have all been developing their own AI chips, which could eventually replace Nvidia's GPUs in their data centers.

The British chip designer Graphcore also claims its graph-processing chips, which process all of the data points across a single graph at once, can process AI tasks more efficiently than Nvidia's GPUs. Therefore, Nvidia's might have established a first-mover's advantage in the AI market, but its long-term dominance isn't guaranteed.

The recent hype regarding ChatGPT, DALL-E, Midjourney, and other popular generative AI platforms might have also created an AI bubble which could be popped by concerns about copyright violations, the proliferation of fake information, and the disruptions of entire industries. Governments could also start to strictly regulate the usage of AI tools to mitigate those long-term risks. If that happens, Nvidia's sales of data-center GPUs (60% of its sales last quarter) could slow down.

The bears also believe a lot of growth has already been baked into Nvidia's stock at 56 times forward earnings and 24 times this year's sales. By comparison, Advanced Micro Devices (NASDAQ: AMD) trades at 39 times forward earnings and 8 times this year's sales. So if Nvidia fails to meet the market's ever-rising expectations for its data-center business, its stock could easily be cut in half. That's probably why Nvidia's insiders sold more than three times as many shares as they bought over the past 12 months.

What the bulls will tell you about Nvidia

The bulls believe that Nvidia will remain far ahead of its potential challengers in the AI chip market for the foreseeable future. For example, they'll point out that even though Google recently claimed its fourth-generation TPUs (tensor processing units) were 1.2 to 1.7 times faster and up to 1.9 times more power efficient than Nvidia's A100 chips, the tech giant didn't directly compare those TPUs to Nvidia's newest H100 chips.

As for Graphcore, it's still an unprofitable start-up that generated a mere $5 million in sales in 2021 and lost its entire AI networking chip team to Meta earlier this year. Therefore, Graphcore's technology might seem promising on paper, but it seems unlikely that it can catch up to Nvidia -- which could generate $43 billion in revenues this year -- anytime soon.

The bulls also don't expect the AI bubble to burst. Instead, they believe it's only the beginning of a new secular trend which will light a long-term fire under Nvidia's data-center business. While new regulations could force companies to adopt a more cautious approach toward adopting new AI tools, the underlying need to crunch more data should remain robust and fuel the market's long-term demand for Nvidia's data-center chips.

If Nvidia can grow its top line at a CAGR of 25% over the next five years as it expands its data-center and gaming businesses, its annual revenue would more than triple. Assuming its valuations hold steady, its stock could triple again. That wouldn't be as impressive as Nvidia's gains from the past year, but it could still easily outperform the broader market.

Which argument makes more sense?

I'm bullish on Nvidia's long-term prospects, but the bears are right about its valuations. Nvidia's stock is priced for perfection right now, so it will be punished harshly for even the smallest setbacks or guidance reductions. Rising interest rates could also exacerbate that pain by compressing Nvidia's valuations again. Therefore, investors can certainly nibble on Nvidia at these levels, but they shouldn't accumulate a large position until its valuations cool off to more sustainable levels.