The AI stock crash of January 2025 was triggered by Chinese startup DeepSeek’s release of its R1 model on January 20, which claimed performance comparable to leading AI models at a fraction of the cost. Nvidia alone lost $589 billion in market value on January 27, 2025—the largest single-day loss in U.S. stock market history. The crash stemmed from investor fears that AI infrastructure spending might slow dramatically if cheaper, more efficient alternatives could deliver similar results without expensive hardware.

What Triggered the Sudden AI Stock Market Decline?

DeepSeek’s announcement that it developed its R1 model for under $6 million using less powerful chips sent shockwaves through tech markets. This contrasted sharply with OpenAI and Google spending billions on computational resources. The Nasdaq fell 3.1% on January 27, with the Philadelphia Semiconductor Index dropping 9.2%. Investors questioned whether the massive capital expenditures by Microsoft, Meta, and Google on AI infrastructure were justified if cheaper methods could achieve comparable results.

Which AI Companies Were Most Affected by the Crash?

Nvidia bore the brunt, with shares plummeting 17% in a single day. Other semiconductor stocks followed: Broadcom dropped 17.4%, ASML fell 7.6%, and AMD declined 6.2%. Cloud providers also suffered—Microsoft fell 2.1% and Amazon dropped 2.3%. The crash erased approximately $1 trillion in combined market value from AI-related stocks within 48 hours, according to Bloomberg data. Taiwan Semiconductor Manufacturing Company (TSMC) lost $90 billion in valuation.

Will the AI Investment Boom Continue Despite This Setback?

Most analysts view this as a temporary correction rather than an AI bubble burst. Major tech companies have reaffirmed their infrastructure spending commitments, with Microsoft CEO Satya Nadella stating that demand for AI compute remains “insatiable.” However, the crash introduced new scrutiny around capital efficiency and return on investment for AI projects, likely leading to more measured growth expectations going forward.