Here's the thing about competing with Nvidia (NVDA): you don't actually have to beat them. At least that's the take from Shay Boloor, Chief Market Strategist at Futurum Equities, who laid out an interesting roadmap for Advanced Micro Devices Inc. (AMD) on Tuesday.

Boloor identified three critical factors that could make or break AMD's 2026, and none of them involve going toe-to-toe with the AI chip giant in a winner-takes-all battle.

Factor One: The CPU Cash Machine

First up is something that might surprise people obsessed with the GPU wars: AMD's CPUs. Boloor calls these data center processors the company's "cash engine," and for good reason. These chips are set to be a major growth driver, benefiting from a perfect storm of tight supply conditions, robust hyperscaler demand, and the possibility of price hikes. With server CPU growth expected to exceed 50%, these processors provide crucial operating leverage as AMD's GPU business scales up. Think of it as the steady revenue stream that funds the flashier AI chip ambitions.

Factor Two: The Platform Play

The second factor revolves around something called Helios, which is essentially a real-world stress test. This isn't just about shipping a few chips to see if they work. It's about sustained volumes with actual customer deployments that will prove whether AMD can transform from being just another component supplier into a full platform vendor. That's a massive distinction in the tech world. Components are commodities; platforms are ecosystems.

Factor Three: Changing the Game, Not Winning It

Here's where Boloor's analysis gets really interesting. AMD doesn't need to "beat" Nvidia, he argues. Instead, it needs major customers like Microsoft (MSFT), Meta Platforms (META), Oracle Corporation (ORCL), and Tesla (TSLA) to fundamentally change their procurement rules.

The goal isn't to win every deal against Nvidia. It's to make multi-vendor AI infrastructure the default approach, avoiding single-vendor dependency. If AMD becomes part of the standard architecture rather than fighting for each contract deal-by-deal, that's a win. It's about being the automatic second choice that becomes mandatory for diversification, not the underdog constantly trying to steal market share.

Tariffs and Competitive Moves

This strategic framework comes at an interesting moment. President Donald Trump just imposed a 25% tariff on select high-end AI chips from companies including Nvidia and AMD on Wednesday. The move is part of a broader push to reduce U.S. dependence on foreign-made chips, which policymakers view as posing economic and security risks.

But AMD isn't sitting still. The company has been aggressively rolling out new chips and faster systems aimed squarely at the data center market, directly challenging Nvidia's dominance. CEO Lisa Su has been particularly bullish about the future, predicting a massive increase in global compute demand that extends far beyond traditional data centers. She believes we're entering what she calls the "yottascale" era of computing, which would require an unprecedented expansion of global computing capacity. For context, a yottabyte is a trillion terabytes. We're talking about computational needs at a scale that's hard to even conceptualize.

The Numbers Tell a Story

AMD's recent performance suggests investors are buying into the vision. Over the past year, the stock has climbed 86.40%, closing Wednesday at $223.60 after edging 1.19% higher. The company ranks in the 85th percentile for quality and the 91st percentile for momentum, reflecting strong performance across both fundamental and technical measures.

The strategic question for AMD isn't whether it can dethrone Nvidia. It's whether it can execute on these three fronts: maximizing the CPU cash engine, proving it can deliver complete platform solutions, and convincing the world's biggest tech companies that single-vendor AI infrastructure is too risky. Get those right, and AMD doesn't need to be number one to be wildly successful.