IBM’s Historic Stock Rout Exposes a New Winner in the AI Spending Race: Micron
IBM’s worst trading day in nearly six decades delivered an unexpected signal to investors: the artificial intelligence boom is redirecting corporate technology budgets away from traditional software and services and towards scarce memory chips and computing infrastructure.
IBM shares plunged 25.2% on July 14, their steepest one-day decline since at least 1968, after the company issued an unusual preliminary earnings warning. The sell-off erased roughly $67 billion from the technology group’s market value in a single session.
The warning initially appeared to be an IBM-specific setback, driven by weaker-than-expected second-quarter revenue and delayed contracts. Yet the company’s explanation pointed to a broader shift that could strengthen the position of memory-chip suppliers including Micron Technology.
IBM said customers had abruptly reprioritised capital expenditure towards servers, storage systems and memory components, seeking to secure supply before anticipated price increases. That left less money available for some software, consulting and mainframe projects during the quarter.
Chief Executive Arvind Krishna acknowledged that IBM had failed to anticipate the scale of the spending shift and said several major transactions did not close within the expected period.
Preliminary second-quarter revenue reached approximately $17.2 billion, below Wall Street’s estimate of about $17.9 billion. Infrastructure revenue declined by 7%, while adjusted earnings of $2.93 per share also missed market expectations.
The company’s difficulties illustrate how the economics of AI are changing corporate purchasing decisions. Businesses deploying generative AI systems require large quantities of servers, storage and high-performance memory before they can expand software applications and consulting programmes around them.
As component prices rise and supply remains constrained, customers may bring forward hardware purchases while delaying spending elsewhere.
That dynamic represents a potential benefit for Micron, one of the world’s largest manufacturers of DRAM, NAND flash and high-bandwidth memory used in AI accelerators and data centres.
Micron has said that supply-and-demand conditions for both DRAM and NAND are expected to remain tight beyond 2027, supported by rapid growth in AI infrastructure and rising memory content across servers, devices, vehicles and robotics.
The company has also expanded its exposure to frontier AI development. In June, Micron announced a strategic agreement with Anthropic covering memory and storage architecture, product supply, enterprise adoption of the Claude model and an investment in the AI startup.
IBM’s warning therefore offered investors evidence that companies are not necessarily reducing overall technology investment. Instead, they are shifting a greater share of their budgets towards the physical infrastructure required to run AI workloads.
This distinction is important for the wider technology market. Software companies may face postponed contracts and slower growth even as semiconductor and memory suppliers continue to benefit from urgent infrastructure spending.
The advantage for Micron is not without risks. A rush to secure memory and servers can bring future demand forward, creating the possibility of weaker orders after customers have accumulated sufficient inventory.
Micron’s shares have also experienced significant volatility following a powerful AI-driven rally. Concerns about future supply growth, high valuations and the possibility of reduced capital expenditure by large technology companies have recently placed pressure on memory stocks.
For now, however, IBM’s earnings shock supports the argument that memory remains one of the most strategically important and supply-constrained parts of the AI ecosystem.
The episode also reveals a widening divide inside the technology sector. Companies selling the chips, memory and storage needed to build AI infrastructure may capture spending immediately, while businesses focused on software and services could be forced to wait until customers complete the first stage of their hardware investments.
IBM is expected to release its final second-quarter results and updated guidance on July 22. Investors will be watching closely to determine whether the shortfall was primarily a timing issue or evidence of a more lasting disruption to its business model.
For Micron, the message from IBM’s collapse was considerably more encouraging: even when AI spending damages one established technology company, the same dollars may simply be moving towards another part of the industry.