AMD (NASDAQ:AMD) plans to substantially increase chip supply in 2027 as the semiconductor company seeks to meet rising demand associated with artificial intelligence, Chief Executive Lisa Su said on Tuesday.
AMD has been expanding its AI chip business, including graphics processing units used for AI workloads.
“We’ve been able to increase our supply as we’ve gone through 2026, and we’re going to substantially increase our supply in 2027,” Su told reporters in Taipei.
Su said part of the purpose of her visit to Taiwan was to ensure AMD’s supply chain could increase CPU and GPU production to meet anticipated AI demand.
AMD Holds Supply Chain Discussions in Asia
During the Taiwan visit, Su met Foxconn and was scheduled to meet TSMC as AMD held discussions with local supply-chain partners.
AMD is also working with memory-chip manufacturers to secure sufficient supplies, Su said ahead of a visit to South Korea. The country is home to Samsung and SK Hynix.
Asked whether TSMC had discussed a potential investment in Texas with AMD, Su did not directly address the question. She instead said AMD expected to require additional advanced wafer capacity.
“There’s very, very high demand for the next several years. We can see very high demand, and so we need more advanced wafer capacity,” Su said.
She added that AMD was now planning its capacity requirements three to five years in advance.
Reuters reported last week that TSMC was evaluating a potential investment in Texas. The supplied material does not indicate that an investment decision has been made.
AMD CEO Comments on AI Development and Safety
Su also addressed the broader development of artificial intelligence, saying AI could be “incredibly good for the world” while noting that companies needed to consider associated risks and cooperate on safety.
She said coordination between major technology companies would be important as AI systems develop and described last week’s AI summit involving U.S. President Donald Trump and technology executives as an important step.
Companies attending the meeting agreed to work with “independent auditors” to assess whether AI systems operate as intended by their designers, according to a copy of the agreement cited in the supplied material.