Goldman Sachs is discussing the current software industry in parallel with the newspaper industry, which was disrupted by the Internet in the early 2000s, and the tobacco industry, which was hit hard by regulation in the late 1990s. Goldman Sachs believes that the current valuation decline does not reflect short-term profit fluctuations, but a fundamental doubt about whether the long-term growth and profit margins of the software industry are still valid. Only when earnings expectations truly stabilize can the stock price complete its bottoming process.
When Wall Street starts using the “newspaper industry” to describe software stocks, the market’s fear of AI impact has entered an extreme stage.
In its latest report, Goldman Sachs analyst Ben Snider and his team rarely discussed the current software industry in parallel with the newspaper industry, which was disrupted by the Internet in the early 2000s, and the tobacco industry, which was hit hard by regulation in the late 1990s. This analogy itself is enough to illustrate Wall Street’s pricing of the “AI impact on the software business model.”
Goldman Sachs believes that the current valuation decline does not reflect short-term profit fluctuations, but a fundamental doubt about whether the long-term growth and profit margins of the software industry are still valid. Goldman Sachs reminds that when an industry is identified by the market as facing disruptive risks, the bottoming of the stock price depends on whether earnings expectations are stable, rather than whether the valuation is cheap enough.
From “AI dividend” to “AI threat”: Software stocks are undergoing a collective revaluation.
Goldman Sachs pointed out that in the past week, software stocks have become the “eye of the storm” in the AI impact narrative. The software sector has plummeted 15.00% in a week, and has cumulatively fallen 29.00% from its high in September 2025. The “GS AI at Risk” basket compiled by Goldman Sachs has fallen 12.00% since the beginning of the year.
The direct catalysts that triggered the shift in market sentiment include the release of Anthropic’s Claude collaboration plug-in and the launch of Google’s Genie 3 model. In the eyes of investors, these developments are no longer just “improving productivity”, but are beginning to directly threaten the pricing power, moat and even the value of software companies.
Goldman Sachs clearly pointed out in the report that the current market discussion is no longer just about profit revisions, but whether “the software industry is facing a long-term recession path similar to that of newspapers.”
Valuation seems to be “returning to rationality”, but the market is betting on growth collapse.
On the surface, the valuation of software stocks has fallen significantly: the forward price-to-earnings ratio of the software sector has fallen from about 35 times at the end of 2025 to about 20 times currently, which is at a low level since 2014; the valuation premium relative to the S&P 500 has also fallen to the lowest level in more than a decade.
However, Goldman Sachs emphasized that the problem is not the valuation, but the assumptions behind the valuation are collapsing. The report shows that the profit margins and consensus expected revenue growth of the software industry are still at their highest levels in at least 20 years, significantly higher than the average level of the S&P 500.
This means that the valuation cut given by the market implies expectations of a significant downward revision in future growth and profit margins. Goldman Sachs found through horizontal comparison that in September 2025, when software stocks were still at 36 times P/E, the corresponding medium-term revenue growth expectation was 15.00%-20.00%; while the current valuation of about 20 times corresponds to a growth assumption that has fallen to the 5.00%-10.00% range. In other words, the market is pricing in advance for a “growth cliff.”
The warning of the “newspaper moment”: Valuation is not the bottom, profit stability is.
The most attention-grabbing part of this report is Goldman Sachs’s citation of historical cases. Goldman Sachs reviewed and pointed out that the average stock price of the newspaper industry fell 95.00% between 2002 and 2009, and the real bottoming did not occur when the macroeconomy improved or the valuation was cheap enough, but after the consensus expected profit stopped being revised downward.
A similar situation also occurred in the tobacco industry in the late 1990s: before the Master Settlement Agreement was reached and regulatory uncertainty was eliminated, even if the valuation had been significantly compressed, the stock price continued to be under pressure.
Based on these cases, Goldman Sachs’s conclusion is quite calm and even pessimistic: even if the short-term financial reports show resilience, it is not enough to deny the long-term downside risks brought by AI.
Funds have already voted with their feet: stay away from “AI risk” and embrace the “real economy.”
In the context of rising AI uncertainty, market preferences are shifting from staying away from “AI risk” to embracing the “real economy.”
Goldman Sachs data shows that hedge funds have recently significantly reduced their exposure to the software sector, although they still maintain a net long position overall; while large mutual funds have begun to systematically underweight software stocks in the middle of last year.
At the same time, funds are clearly flowing to sectors that are considered to have “lower AI impact”, including typical cyclical industries such as industry, energy, chemicals, transportation and banking. Goldman Sachs pointed out that its tracked Value factor and industrial cycle-related portfolios have both significantly outperformed recently.
Although the overall tone is cautious, Goldman Sachs has not turned to a fully bearish view. Its analyst team believes that some sub-sectors are still defensive: vertical software is less likely to be directly replaced by AI due to its deep embedding in industry processes and high customer migration costs; information services and business service companies with proprietary data and clear industry barriers may have overestimated the AI impact by the market; some companies that are highly related to software but whose business models are not purely software have recently shown signs of being “wrongly killed.”
But the premise is still clear: only when earnings expectations truly stabilize can the stock price complete its bottoming process.
If the core narrative of software stocks in the past two years was “AI will amplify growth”, then this report from Goldman Sachs marks a turning point – the market is beginning to seriously discuss: whether AI will erode the commercial value of software itself. The real question is not whether software stocks can rebound, but which software companies can prove that they will not become the next newspaper industry.
[Zhao Ying]
Goldman Sachs’ “Newspaper Moment” Warning: How AI Disruption Could Reshape Crypto Valuations
Wall Street’s pessimism has reached a critical juncture, with Goldman Sachs drawing explicit parallels between today’s software industry and the newspaper industry decimated by the internet two decades ago. This isn’t just another market correction—it’s a fundamental reassessment of whether AI represents an enhancement or existential threat to established business models. For crypto investors, this paradigm shift carries profound implications that extend far beyond traditional software markets.
The Newspaper Parallel: A Dire Warning for Digital Assets
When Wall Street’s premier investment bank explicitly compares software to newspapers—a sector that lost 95% of its value between 2002 and 2009—we’re witnessing an inflection point in market psychology. The analogy isn’t accidental; it suggests that Goldman Sachs believes AI may not just augment software companies but potentially undermine their core value propositions.
This has direct parallels in the crypto space. Many blockchain projects position themselves as “software solutions”—whether through DeFi protocols, NFT marketplaces, or enterprise blockchain applications. If AI can deliver equivalent functionality at lower costs or with greater efficiency, these projects could face the same existential threat that newspapers faced from free online content.
The market’s reaction has been swift and brutal. Goldman’s “AI at Risk” basket has plummeted 12% year-to-date, while the broader software sector has shed 29% from its September 2025 peak. In crypto, we’re already seeing similar patterns, with AI-related tokens like AGIX, FET, and RNDR experiencing significant corrections as the narrative shifts from “AI synergy” to “AI competition.”
Valuation Collapse: Not Just Multiple Compression, but Growth Doubt
What’s particularly striking is that the software sector’s forward P/E ratio has contracted from 35x to 20x—yet the market remains unconvinced. This isn’t a simple valuation adjustment; it’s a collapse in growth assumptions. Where the market once priced in 15-20% revenue growth for software companies, it now expects just 5-10%.
For crypto projects, this creates a dangerous dynamic. Many tokens are valued based on future adoption and revenue growth projections. If AI alternatives emerge that can replicate blockchain functionality more efficiently, the growth assumptions underpinning crypto valuations could face similar downward revisions.
Consider DeFi protocols: if AI can provide equivalent financial services with lower overhead and greater user experience, why would users choose decentralized alternatives? This question is particularly acute for protocols that haven’t established strong network effects or unique value propositions beyond decentralization.
Market Rotation: From Digital Dreams to Real Assets
Perhaps most telling is the market’s rotation away from software toward “real economy” sectors less susceptible to AI disruption. In crypto, this could manifest as a renewed focus on:
-
Real World Assets (RWAs): Tokenized assets with tangible value—real estate, commodities, physical infrastructure—may outperform purely digital assets.
-
Privacy and Identity Solutions: As AI blurs the lines between human and machine content, projects that verify authenticity and protect digital identities could gain prominence.
-
Hardware-Dependent Blockchains: Networks requiring specialized hardware or physical infrastructure might benefit from being less susceptible to pure software disruption.
-
Regulatory-Compliant DeFi: Projects proactively engaging with regulatory frameworks may be viewed as more sustainable alternatives to purely permissionless systems.
The smart money is already moving. While hedge funds have reduced software exposure, mutual funds have been systematically underweighting the sector since mid-2025. In crypto, we’re seeing similar shifts, with capital flowing toward infrastructure projects and those with clear revenue streams beyond token speculation.
The AI Paradox: Threat and Opportunity in Crypto
Ironically, while AI threatens many software business models, it simultaneously creates unique opportunities for blockchain technology:
-
AI Verification: As AI-generated content becomes indistinguishable from human-created work, blockchain-based verification systems could become essential for establishing provenance and authenticity.
-
AI Data Markets: Blockchain could facilitate transparent, audited markets for training data, addressing growing concerns about AI data provenance and privacy.
-
Decentralized AI Compute: Projects that democratize access to computational resources could counterbalance the concentration of AI power in corporate hands.
-
AI-Human Collaboration Frameworks: Blockchain protocols that govern how humans and AI interact could become increasingly valuable as these relationships become more complex.
The key differentiator will be whether crypto projects can establish moats that AI cannot easily penetrate. This might involve network effects, specialized hardware requirements, regulatory advantages, or unique combinations of human and machine intelligence.
Defensive Postures: Which Crypto Projects May Withstand the Storm?
Not all crypto projects are equally vulnerable. Based on Goldman’s analysis, certain characteristics may provide relative resilience:
-
Embedded Vertical Solutions: Projects deeply integrated into specific industry workflows—like supply chain management or healthcare records—may face lower disruption risk due to switching costs and regulatory compliance requirements.
-
Proprietary Data Networks: Blockchain ecosystems that aggregate unique, difficult-to-replicate datasets could maintain value even if generic AI becomes more powerful.
-
Hybrid Physical-Digital Models: Projects bridging the physical and digital worlds—like IoT-integrated blockchains or tokenized real assets—may benefit from tangible value anchors.
-
Governance and Coordination Mechanisms: Projects that facilitate complex human coordination where AI cannot easily substitute—like decentralized autonomous organizations with nuanced decision-making requirements.
The Path to Stabilization: When Will the Bleeding Stop?
Goldman’s historical analysis offers a sobering lesson: newspaper stocks didn’t find their bottom until earnings expectations stabilized, not when valuations became attractive. This suggests that crypto markets may not bottom until there’s clarity on how AI will ultimately impact adoption and monetization.
For investors, this means patience may be required. The current valuation compression in crypto could continue until we see:
- Clear evidence that AI complements rather than replaces blockchain functionality
- Sustainable business models that generate revenue independent of token speculation
- Regulatory clarity that establishes blockchain’s role in AI-driven economies
- Network effects that become self-reinforcing regardless of technological alternatives
Conclusion: A Narrative Shift in Crypto Markets
Goldman Sachs’ “newspaper moment” warning represents more than just a traditional market call—it’s a fundamental challenge to the digital asset thesis. As AI capabilities advance, crypto projects must answer a critical question: Are we building technologies that will be enhanced by AI, or simply replaced by it?
The coming months will likely separate projects that can articulate their unique value in an AI-dominated world from those that face existential threats. For experienced crypto investors, this may be the most important narrative shift since the 2018 bear market—one that will define which projects survive and thrive in the next technological revolution.
The question is no longer whether crypto can benefit from AI, but whether crypto can establish itself as an indispensable complement to—or safeguard against—the AI-driven future.