We are months into the RAM crisis, and the uncomfortable part is the direction of travel. Instead of games getting cheaper, hardware is getting more expensive. IGN notes that Nvidia and AMD have both reportedly raised prices on graphics cards “just this week,” and that shift matters because it directly controls whether players upgrade at all.

This is not your typical scarcity cycle. In earlier hardware crunches, like those tied to cryptocurrency mining, only one or two component types would spike because certain retail channels sold out. The current AI crunch is different. According to Reuters, AI companies made promises of future deals with Samsung, Micron, and SK Hynix, pledging to buy up memory chips before they even existed. Even after “plenty of those lofty non-contracts have fallen through,” memory manufacturers are still funneling a huge portion of capacity to AI and datacenters. The result is that “every single little device that uses memory,” from the PS5 to your car, is going up in price.

If you are an executive watching this from the outside, the supply-chain logic is the key. Memory is not just a line item for gaming PCs. It is a foundational input that leaks into everything: consoles, personal devices, vehicles, and the infrastructure that powers AI workloads. So when memory capacity gets pulled toward datacenters and AI systems, it does not stay neatly contained to “AI hardware” the way some people imagine. It shows up at the consumer end as higher costs, and those costs then become part of the pricing baseline players expect.

IGN also flags what happens when the incentives stay misaligned. There is already a “class action lawsuit” targeting memory companies, alleging price fixing, and while it could take years to reach resolution, the existence of the case itself is a signal. Even if the timeline drags, litigation tends to force companies to spend management attention on legal strategy, documentation, and compliance. That is board-level work, not just courtroom work.

Now zoom out to games themselves. IGN says that when it looked at AI for gaming earlier in 2024, expectations were higher. At CES 2024, the outlet spent time talking with Intel, AMD, and Nvidia about what AI could do for video games. The “utopian visions” included NPCs you could talk to naturally and inventory systems that managed themselves. Fast forward “two and a half years,” and IGN reports that the reality feels less impressive. The most frequent outcome, it says, is the need to check Steam store pages for whether developers used generative AI to add “ugly art” to their worlds.

That does not mean all AI in gaming is useless. IGN points to earlier, arguably more grounded wins. Nvidia launched the RTX 2080 with DLSS, or deep learning super sampling, back in 2018. It also notes that AMD, Intel, and Sony have followed with their own AI upscaling tech. Making games easier to run is framed as a net positive, because performance improvements can increase accessibility.

The problem is escalation. IGN explains that Nvidia added frame generation starting with the RTX 4090 in 2023. Nearly three years later, the technology is described as “divisive,” even as it is said to have grown on the writer. Then in March, IGN says Nvidia pushed DLSS “to its breaking point” with DLSS 5. The presentation, as described by IGN, takes the final output of a rendered frame and then uses it as input to present a new image to the user. It includes an example from Jensen Huang’s presentation where Grace from Resident Evil Requiem looked like she came out of a mobile game ad. Those specifics matter because they illustrate the trade: the more AI is asked to synthesize or approximate, the more players may notice quality seams.

Other companies are layering their own AI features on top. IGN mentions that Microsoft, while IGN was reviewing the Xbox Ally X in October last year, appeared to shove “Gaming Copilot” into the Xbox Game Bar. IGN’s description is blunt: you can hold down the library button, and a laggy AI companion shows “outdated advice” for the game you are currently playing.

For decision-makers, the sting is timing. IGN connects these AI feature rollouts to the period when gaming is becoming prohibitively expensive due to the AI-fueled RAM crisis. The core argument is not that AI upgrades are purely bad, but that players feel like they are paying for features they do not necessarily value, while paying more for the privilege of playing at all. IGN even notes that the writer knows the economics do not work exactly like the gut-feel suggests, but the customer perception is what drives adoption, churn, and refunds.

There is also an information and labeling flashpoint. IGN points to Tim Sweeney throwing a fit that labeling AI usage on store pages is irrelevant because, he thinks, AI will simply be used everywhere. The practical question for publishers, platforms, and hardware partners is whether transparency keeps up with reality. If players start to assume AI is in everything, trust can erode. And if trust erodes, “AI features” can flip from differentiator to distraction.

So the headline-level takeaway is the same one the market is starting to feel in its wallet: the AI boom is not just changing compute, it is reordering supply chains, tightening memory allocation toward AI and datacenters, and pushing consumer hardware costs higher. Until that reverses, games may not get more affordable, and AI features may remain uneven in value. For executives across gaming, hardware, and platform businesses, the strategic stake is clear: you are selling experiences into a world where the base cost of playing keeps climbing, while product quality and transparency narratives struggle to keep pace.