AMD plans to substantially increase chip supply in 2027 as demand for artificial intelligence hardware continues to grow. Still, the move may not immediately translate into cheaper laptops, gaming PCs or consoles. Chief executive Lisa Su said the company is working with manufacturing and supply-chain partners to expand production as demand for computing capacity remains higher than available supply.
The announcement comes as AMD continues to expand its position in the rapidly growing AI accelerator market. The company is competing with Nvidia for business from data centres and other organisations deploying large-scale AI systems, creating additional pressure on the semiconductor supply chain.
AMD prepares for another major increase in demand
Speaking to reporters during a visit to Taiwan, Su said AMD had already increased its supply during 2026 but expected to go considerably further next year. “We’ve been able to increase our supply as we’ve gone through 2026, and we’re going to increase our supply in 2027 substantially,” she said.
The chief executive made the comments while meeting suppliers and manufacturing partners involved in AMD’s chip production. Her visit included discussions with companies such as Foxconn and TSMC, aimed at ensuring the wider supply chain can expand production of CPUs, GPUs and other computing products needed to meet AI demand.
Su said AMD is now planning further ahead than it did in the past because demand for computing capacity is expected to remain very high. The company has moved from planning one or two years to working with partners on capacity requirements three to five years out.
The expansion is not limited to semiconductor manufacturing itself. Advanced wafer capacity, packaging and substrates are all important parts of the process, and shortages in any of these areas can restrict the number of finished processors and accelerators that can reach customers.
AI remains the priority for AMD’s expanding capacity
For AMD, the strongest source of demand is increasingly the AI sector. Data centres require large numbers of powerful processors and accelerators to train and operate AI models, while cloud providers are investing heavily in infrastructure to support growing workloads.
That demand is creating a difficult situation for the wider consumer technology market. More manufacturing capacity is positive in principle, but the additional chips AMD and its partners are prioritising are largely tied to the rapidly expanding computing needs of AI businesses.
AMD is also working to secure sufficient memory supplies. Su has indicated that the company is engaging with memory manufacturers as it prepares for further growth, including the high-bandwidth memory required by AI systems. She was also due to meet Samsung and SK Hynix during her visit to South Korea.
The supply-chain push reflects how valuable AI hardware has become for semiconductor companies. AI accelerators can command significant prices, while demand from major technology companies and data-centre operators has created an opportunity for chipmakers to generate substantial growth.
That does not mean AMD is abandoning consumer processors or graphics cards. Ryzen CPUs and Radeon GPUs remain important parts of its business, and the same manufacturing infrastructure can support several product categories. However, the rapid expansion of AI infrastructure means that consumer buyers could continue to face a less favourable supply environment.
Laptop and gaming hardware prices could remain under pressure
The biggest concern for consumers is that increased AMD production does not necessarily mean more affordable hardware. If much of the additional capacity is absorbed by AI-related products, the supply of chips destined for conventional PCs and gaming systems could remain relatively constrained.
That could keep pressure on laptop and PC prices, particularly if memory components also remain expensive. Manufacturers need processors, graphics hardware, and memory to build complete systems, so shortages in one part of the supply chain can affect a device’s final price.
There is also growing concern that strong AI demand could reshape the semiconductor market for years, not months. Su’s decision to plan capacity three to five years ahead suggests AMD expects the demand for advanced computing hardware to remain elevated.
Some consumers have already expressed frustration that the PC market could increasingly become secondary to data-centre demand. One Reddit user described the potential outcome by saying: “Yeah, I have a feeling the consumer market will end up being ‘whatever the data centres consider obsolete’ soon. We might have to contend with buying used data centre CPUs and GPUs soon.”
That scenario is unlikely to become a straightforward replacement for consumer hardware. Data-centre GPUs are designed for very different environments and workloads, and issues involving cooling, physical design, display outputs and software compatibility can prevent them from being practical gaming cards.
For now, AMD’s announcement is therefore a mixed development. More production should help ease some of the severe capacity constraints affecting the semiconductor industry, but the immediate priority is meeting the enormous appetite for AI computing. Consumers may have to wait longer for increased chip production to translate into noticeably better prices or availability for laptops, desktops and gaming hardware.
Su has also acknowledged the broader implications of the AI boom. She described artificial intelligence as potentially “incredibly good for the world” while stressing the importance of recognising its risks and working together on safety. As AMD expands production to meet demand, the balance between AI infrastructure and consumer technology is likely to remain an important issue for the semiconductor industry.
