Goldman Sachs Warns Software Stocks Face AI Disruption Cliff
Goldman Sachs released a report stating that software stocks will face another wave of downside risk.
Goldman Sachs expects that the debate over AI disruption and the uncertainty of terminal values for multiple companies will last for at least several quarters, with the threat of disruption forming a persistent ceiling before the large-scale adoption of AI; it is currently difficult to falsify this narrative, and several quarters or even years of evidence will be needed to show AI's real impact on profits.
Mechanically, institutional investors are shifting funds from overvalued traditional software stocks to AI-native companies and targets with clear profit realization, putting pressure on the SaaS and enterprise software sectors, while leaders in AI infrastructure and application implementation benefit, leading to ongoing revaluation pressure on overall tech valuations.
Source: Public Information
ABAB AI Insight
Goldman Sachs has previously lowered expectations for the software sector multiple times for 2025-2026, and this report continues its cautious judgment on the narrative of AI as "disruptive rather than enhancing," aligning with views from investment banks like Morgan Stanley, emphasizing that the terminal value revaluation cycle will far exceed market optimism.
In terms of capital pathways, software company management continues to increase investments in AI capabilities, but Goldman Sachs points out that most of these are defensive expenditures rather than growth-driven, with resources shifting from traditional subscription renewals and sales teams to AI R&D spending. Strategically, they are attempting to hedge against disruption risks, but it is difficult to quickly form differentiated profit evidence, leading to valuation pressure.
Similar cases include the traditional enterprise software being rapidly replaced by Salesforce and others during the cloud computing transformation in the 2010s, and the current traditional SaaS companies facing growth slowdowns even after implementing copilot features during multiple rounds of AI enthusiasm. The software industry is currently at the threshold of transitioning from the narrative of AI efficiency to the verification of real profits.
Essentially, this is a technological replacement: traditional software business models are gradually being replaced by AI-native architectures. The mechanism is that under the rapid enhancement of AI capabilities, additive copilot features struggle to establish a long-term moat, leading to pricing power shifting from existing SaaS companies to platforms that control core AI models and data loops, while extending the market's verification window for software stock profits.