AMD's AI Chip Renaissance: How the Underdog is Challenging Nvidia's Dominance
- Daniel
- Jun 19
- 7 min read
Toronto – June 19, 2025 The artificial intelligence revolution has created one of the most valuable and fiercely contested markets in modern technology. While Nvidia has dominated this space with an estimated 80-90% market share in AI training chips, Advanced Micro Devices (AMD) is mounting an increasingly credible challenge that could reshape the competitive landscape. With the recent launch of their MI350 series chips and ambitious roadmap extending through 2026, AMD is positioning itself as the primary alternative to Nvidia's AI dominance—a development that could prove transformative for investors willing to bet on the underdog.
The David and Goliath Dynamic
For years, AMD has lived in the shadow of both Intel in CPUs and Nvidia in graphics processing. However, the company's persistence in developing competitive alternatives has begun paying dividends, particularly in the AI acceleration market. The stark reality is that Nvidia's near-monopoly in AI chips has created supply constraints, premium pricing, and a desperate need for viable alternatives among tech giants like Microsoft, Google, and Meta.
This market dynamic has created a unique opportunity for AMD. Companies are actively seeking to reduce their dependence on a single supplier, especially when that supplier can command premium prices due to market dominance. AMD's AI chips would cost less to operate and less to acquire, according to company executives, presenting a compelling value proposition for cost-conscious enterprises.
The Technical Breakthrough: MI350 Series
AMD's latest salvo in the AI chip wars comes in the form of their MI350 series, which represents a quantum leap in performance. The MI355 chips, which started shipping earlier this month, are 35 times faster than predecessors, demonstrating the company's ability to achieve dramatic generational improvements.
The technical specifications are impressive and competitive with Nvidia's offerings. The GPUs will feature up to 432GB of HBM4 and memory speeds of up to 19.6TB per second, positioning them to compete directly with Nvidia's GB300 Blackwell Ultra processors. This represents a 4x, generation-on-generation AI compute increase and a 35x generational leap in inferencing, making AMD's chips genuinely competitive for the first time in the AI acceleration space.
The MI350 series isn't just about raw performance—it's about creating a comprehensive ecosystem. The chips deliver leading memory capacity (288GB HBM3E from Micron and Samsung Electronics), addressing one of the key bottlenecks in AI workloads: memory bandwidth and capacity.
Strategic Partnerships and Market Validation
Perhaps most significantly, AMD has secured validation from one of the most important players in the AI space. OpenAI taps its newest chips, providing crucial market credibility for AMD's AI acceleration efforts. When the company behind ChatGPT chooses your chips, it sends a powerful signal to the broader market about technical competence and reliability.
This partnership extends beyond just chip sales. AMD is building a complete ecosystem around their AI chips, including the upcoming "Helios" server that AMD plans to release next year based on their MI400 series chips. This systems-level approach mirrors Nvidia's successful strategy of selling complete solutions rather than just individual components.
Government Support and Strategic Importance
The U.S. government's CHIPS and Science Act has created a favorable environment for domestic semiconductor development, with close to $30 billion in subsidies for advanced semiconductor manufacturing, aiming to bring cutting-edge artificial-intelligence chip development and manufacturing to American soil. While Intel has received the lion's share of direct funding, AMD benefits from the broader ecosystem development and the government's strategic interest in reducing dependence on foreign chip manufacturing.
The geopolitical dimension adds another layer to AMD's investment thesis. As the U.S. government implements increasingly strict controls on AI chip exports to China and other countries, domestic alternatives like AMD become more strategically valuable. AMD says that the U.S. government's license control requirement for exporting AI chips to China and certain other countries may impact its earnings materially, but this also creates opportunities in friendly markets where AMD can position itself as a reliable, government-approved alternative.
The Competitive Landscape
The AI chip market is experiencing explosive growth, with industry analysts projecting it could exceed $500 billion by 2028. In this context, even capturing a modest percentage of Nvidia's market share could translate to tens of billions in revenue for AMD. The company doesn't need to overtake Nvidia entirely—it simply needs to establish itself as a credible alternative to justify significant investor returns.
AMD launched the MI350X/MI355X GPUs which could be competitive to Nvidia's HGX B200 solutions for inference of small to medium LLMs on a performance per TCO basis. This total cost of ownership advantage is crucial, as many AI applications don't require the absolute highest performance but do require cost-effective scaling across thousands of chips.
Market Skepticism and Opportunity
Despite these technical achievements and strategic partnerships, AMD shares are flat so far in 2025, signaling that Wall Street doesn't yet see it as a major threat to Nvidia's dominance. This market skepticism may actually represent an opportunity for contrarian investors. The stock market often takes time to recognize fundamental shifts in competitive dynamics, particularly in technical markets where the implications of product launches take quarters or years to fully manifest in financial results.
The investment community's cautious stance toward AMD's AI ambitions likely reflects several factors: Nvidia's historical dominance, the technical complexity of AI chip development, and uncertainty about AMD's ability to scale production and support. However, these same factors that create market skepticism also create the potential for significant upside if AMD executes successfully.
The Roadmap Ahead
AMD's ambitions extend well beyond the MI350 series. The company has outlined a clear roadmap with the Instinct MI350 series, to be built on next-generation architecture, set to be released in 2025 while the Instinct MI400 series is planned for 2026. This provides visibility into sustained innovation and competition with Nvidia's roadmap.
The MI400 series represents AMD's next major leap, built on advanced architectures that could further close the performance gap with Nvidia while maintaining cost advantages. The company's ability to maintain this aggressive development pace while scaling production and support will be crucial to their long-term success in the AI market.
Investment Thesis: The Case for AMD
From an investment perspective, AMD presents a compelling asymmetric risk-reward profile in the AI chip market. The company trades at a significant discount to Nvidia while offering exposure to the same underlying AI growth trends. Several factors support the bull case:
Market Diversification Imperative: Large tech companies are actively seeking alternatives to reduce Nvidia dependence, creating natural demand for competitive alternatives.
Cost Advantage: AMD's positioning as a lower-cost alternative with competitive performance could capture significant market share in price-sensitive applications.
Technical Credibility: The MI350 series demonstrates that AMD can develop genuinely competitive AI chips, not just "good enough" alternatives.
Ecosystem Development: AMD is building comprehensive solutions including hardware, software, and cloud services, creating stickiness and higher margins.
Government Support: Indirect benefits from CHIPS Act funding and strategic importance as a domestic alternative.
Risks and Challenges
However, significant challenges remain. Nvidia's software ecosystem, particularly CUDA, creates substantial switching costs for developers. AMD's ROCm platform is improving but still lacks the maturity and third-party support of Nvidia's development tools. Additionally, Nvidia's pace of innovation shows no signs of slowing, and the company's massive R&D budget provides significant competitive advantages.
Production scaling presents another challenge. Even if AMD's chips are competitive, the company must prove it can manufacture them at scale while maintaining quality and cost advantages. The semiconductor industry is littered with technically superior products that failed due to manufacturing or supply chain issues.
Here’s a deeper dive into AMD’s valuation on a P/E basis:
Trailing P/E (~TTM)
≈ 93× based on trailing twelve-month EPS .
Industry context: Semiconductor sector median is around 29×, with AMD significantly above .
Peer comparison: AMD’s P/E (~92 – 93×) vs peer average (~69×) implies it’s priced at a premium .
Forward P/E
Currently about 34×, which is much more reasonable and reflects expected earnings growth .
A forward P/E in the mid-30s is fairly typical for a growth leader in semiconductors.
PEG Ratio & Growth Prospects
PEG (5‑yr expected) is under 1 (~0.7), signaling AMD’s fast growth may justify its premium valuation .
Analyst Insights & Relative Valuation
AlphaSpread estimates an intrinsic value of US $140, about 10% above the current ~$127 price .
Morningstar/Barron’s describe AMD as “unloved AI bargain” trading below fair value and marking it as a top alternative in the AI hardware space .
On the flip side, some value investors caution that it “only looks undervalued… in a tech & AI bubble” and would prefer waiting for a dip toward ~$100–86 .
Verdict: Buy, Sell, or Hold?
Aspect | Viewpoint |
Trailing P/E | Overvalued relative to peers—suggests caution |
Forward P/E & PEG | Reasonable when factoring growth—supports bullish case |
Analyst Targets | Intrinsic/fair estimates (~$130–140) ~ close to current price |
Risk Appetite | High multiple exposes shares to downside if growth slows |
Buy if you’re confident in AMD’s AI/server momentum and growth engine.
Hold if you own shares: valuation is roughly in line with fair value; upside is moderate while downside exists in case of slower growth.
Wait/Sell if you prefer value plays—some argue a pullback to $100+ could be a better entry point.
Final Take
AMD’s trailing P/E is high, but its forward P/E and PEG signal strong growth expectations. With analysts eyeing fair values near or just above today’s price and mixed sentiment among investors, a neutral-to-cautious “Hold” stance seems prudent—unless you’re truly bullish on continued AI-market share gains, in which case adding could be justified.
Conclusion: A Calculated Bet on Competition
AMD's emergence as a credible competitor in AI chips represents one of the most significant developments in the semiconductor industry in years. The MI350 series launch demonstrates technical competence, while partnerships with companies like OpenAI provide market validation. The company's aggressive roadmap through 2026 suggests sustained competition with Nvidia rather than a one-time product push.
For investors, AMD offers exposure to AI growth trends at a significantly lower valuation than Nvidia, with the potential for dramatic outperformance if the company successfully captures even a modest portion of the AI chip market. The current market skepticism may represent an opportunity for patient investors willing to bet on competition eventually eroding monopolistic market positions.
The AI revolution is still in its early stages, and the market is large enough to support multiple winners. While Nvidia's dominance is unlikely to disappear overnight, AMD's technical progress and strategic positioning suggest that the AI chip market is evolving from a monopoly toward a more competitive landscape. For investors seeking exposure to AI growth with potentially superior risk-adjusted returns, AMD's emerging competitive position merits serious consideration.
The key question isn't whether AMD will overtake Nvidia—it's whether the company can establish itself as a sustainable competitor in a market large enough to reward multiple players. Based on recent developments, that outcome appears increasingly likely, making AMD a compelling investment opportunity for those willing to bet on the power of competition in technology markets.
So are you buying, holding or selling?
Check out my other blog posts here
Comments