We are in the "mid-game" of a great revolution, where all extreme optimism and extreme panic are attempts to prematurely overdraw the yet-to-come conclusion.
Author: market participant
Translation: Deep Tide TechFlow
Deep Tide Introduction: As the new wave of AI agents led by OpenClaw and Claude Code sweeps through social media, the author keenly senses a frenzy reminiscent of the NFT era in 2021.
This article analyzes how social media amplifies technological narratives, how Wall Street indiscriminately sells off due to the bias of "AI killing software," and why giants like Salesforce and ServiceNow are still misjudged by the market despite delivering astonishing results.
The author believes we are in the "mid-game" of a great revolution, where all extreme optimism and extreme panic are attempts to prematurely overdraw the yet-to-come conclusion.
The full text is as follows:
This wave of excitement around OpenClaw and Claude Code reminds me of the hysteria of the NFT era.
The emergence of new technologies comes with practicality while resonating culturally and narratively with the spirit of the times. Like every technology that captures the collective imagination at the right moment, it is being processed through the same "distortion machine"—the very machine that turned JPEG images of monkeys into a $40 billion asset class.
The pattern is the same: true innovation arrives, early adopters discover real value. Then, the social layer takes over everything—suddenly, the conversation shifts away from the technology itself and becomes a performance about "taking sides."
Claiming "this is the future" becomes a badge of honor among insiders. Writing guides, deep thinking pieces, and exaggerating the current value can earn social validation. The compounding speed of opinions even surpasses the technology itself.
(I promise there will be a viewpoint on the financial markets later).
Cognitive Distortion Machine
X makes things worse. Social media is increasingly seen as a legitimate lens on reality, yet it distorts the image of facts.
The loudest voices are not representative—they are performing "firm beliefs" for an audience that rewards such behavior. Every mainstream platform runs on engagement, and engagement rewards extremes. "This is interesting and useful" won’t be widely shared, but "this changes everything, your job is at risk" will.
A hundred retweets saying "this changes everything" are not a signal but an echo. Echoes are mistaken for consensus, consensus is mistaken for truth, and truth is mistaken for investable theory.
If Girard were to see this scene, he would surely shine. When enough people perform "faith performances" about a certain outcome, the performance itself becomes confused with evidence supporting that outcome. The NFT era proved this definitively: people didn’t want JPEGs; they wanted "what others want" [1].
What is Real?
The latest model capabilities are astonishing—far more impressive than NFTs, which have almost no practical capabilities beyond speculation and cultural signaling.
I use these tools every day. They enhance my efficiency in concrete, measurable ways. The underlying models are indeed impressive, and the trajectory of improvement is steep. When I compare what I could do with these tools six months ago to today, the incremental progress is enormous.
Moreover, the broader potential is limitless. AI-assisted programming, research, analysis, writing—these are not hypothetical use cases; they are happening and creating real value for those who use them wisely.
I don’t want to be the person in 1998 who scorned the internet. That’s not the point; I am very bullish on AI in the long term. The focus is on the timeline and the gap between potential and reality.
What is Still Not Real
No—Claude will not immediately catalyze social upheaval. This does not mean humans no longer need interfaces to manage work. It also does not mean Anthropic has won the AI war.
Consider what those most breathless viewpoints actually require you to believe: that enterprise software—decades of accumulated workflows, integrations, compliance frameworks, and institutional knowledge—will be replaced in a few quarters rather than years? That per-seat billing models will die overnight? That companies with over $10 billion in annual revenue and 80% gross margins will evaporate because a chatbot can write a function? [2]
Dan Ives of Wedbush pointedly states: "Companies will not completely overturn hundreds of billions of dollars in software infrastructure investments to migrate to companies like Anthropic and OpenAI" [3]. And Jensen Huang, who has more reason than anyone to tout AI's disruptive power, calls the concept of "AI killing software" "the most illogical thing in the world" [4].
Those most actively declaring an "endgame" (thanks to @WillManidis for popularizing this term) are often those who stand to benefit the most from your "unwavering belief": followers, consulting contracts, subscription fees, conference invitations. The incentive structure rewards bold predictions that bear no responsibility for timing.
The Market's Mirror
What’s interesting to me is that the market is making the same mistake on the other side of the table.
When Anthropic released its Claude Cowork plugin on January 30, less than a week later, $285 billion evaporated from software, financial services, and asset management stocks [5].
The software ETF—$IGV—has fallen 22% this year, while the S&P 500 is rising. Out of 110 constituents, 100 are in the red. The RSI index hit 16, the lowest reading since September 2001 [6].
Hedge funds are aggressively shorting software stocks and continuously adding to their positions [7]. The narrative logic is: AI kills SaaS (Software as a Service). Every per-seat billed software company is a "walking dead."
This sell-off is indiscriminate. Companies with completely different risk profiles affected by AI are treated as the same trading benchmark [8]. When 100 out of 110 names in the index are down, the market is no longer analyzing; it is indulging in the climax of the narrative.
Note: Since I began writing this piece, a recovery may have already started.
Throwing Out the Bathwater, Losing the Baby
Let’s look at what’s actually happening inside those companies deemed to be facing extinction.
Salesforce's Agentforce revenue grew 330% year-over-year, with an annualized revenue exceeding $500 million, generating $12.4 billion in free cash flow. The forward P/E ratio is 15 times. They just announced a revenue target of $60 billion for fiscal year 2030 [9]. This is not a company being disrupted by AI—this is a company building the AI enterprise delivery layer.
ServiceNow's subscription business grew 21%, operating profit margin expanded to 31%, and they authorized a $5 billion stock buyback. Their AI suite, Now Assist, reached an annual contract value (ACV) of $600 million, with a goal to exceed $1 billion by year-end [10]. However, its stock price has fallen 50% from its peak.
Should these names be moderately repriced due to risk? Perhaps. But smart people have been pricing this in for years. As many smarter than I have pointed out: this sell-off requires you to believe both that "AI capital expenditures are collapsing" and that "AI is powerful enough to destroy the entire software industry" [11]. Both cannot be true at the same time. Pick one.
Identifying Real Risks
Will some companies be truly replaced? Yes.
Tooling solutions that provide standardized single workflows (Point solutions) are vulnerable. If your entire product is just an interface layer built on non-proprietary data, you’re in trouble. LegalZoom dropped 20%—for companies like this, concerns are substantive [12]. When AI plugins can automatically conduct contract reviews and classify NDAs, the value proposition of paying traditional vendors for the same functionality becomes hard to defend.
But companies with deep integrations, proprietary data, and platform-level foundations are an entirely different story. Salesforce is deeply embedded in the tech stacks of every Fortune 500 company. ServiceNow is the system of record for enterprise IT. Datadog's consumption-based model means more AI compute will directly translate into more monitoring revenue—their non-AI business has actually accelerated to a year-over-year growth of 20% [13].
Selling off digital infrastructure because "AI kills software" is as absurd as selling off construction equipment stocks because buildings are going up.
We’ve Been Through This
The SaaS crash of 2022 is instructive. The sector fell over 50%. The median forward revenue multiple dropped from 25 times to 7 times—below pre-pandemic levels [14]. Yet during that time, earnings reports were consistently strong. The subsequent rebound was significant—the Nasdaq rose 43% in 2023. Admittedly, the catalysts at that time were more about interest rate shocks than fundamental deterioration.
The DeepSeek panic in January 2025 is closer. Nvidia plummeted over concerns that cheap Chinese AI models would render the entire AI infrastructure buildout meaningless, but then fully recovered [15]. That fear was structurally similar to today: a single product release triggered a survival crisis-style reassessment of the entire industry.
Many observers have drawn direct comparisons between the current moment and the early stages of the internet bubble burst—tech stocks fell while consumer staples, utilities, and healthcare stocks rose [16]. But one thing about the internet bubble burst: Amazon fell 94%, then became one of the most important companies in the world. The market tries to price the "endgame" while the game is still in progress, creating one of the greatest buying opportunities in history.
Jim Reid of Deutsche Bank said a profound truth: "At this stage, identifying long-term winners and losers is almost pure guesswork" [17].
I bet he’s right. And this uncertainty—acknowledging that we still don’t know how it will end—is precisely why this indiscriminate sell-off is misguided.
The Endgame Fallacy
The hype merchants on X and the panic sellers on Wall Street are making the same mistake on opposite ends of the board.
One group says AI has already won, the future is here, and all institutions and job functions will be rewritten from now on. The other group says AI has already killed software, subscription revenue is dead, and $10 billion in free cash flow doesn’t matter because the business model is obsolete.
Both sides jumped to the "endgame" when there are still many moves to be made in the game. The gap between our current situation and the technological vision will be filled by chaotic, incremental, company-specific progress. Some software companies will integrate AI and become stronger; a few will truly be replaced; most will adapt—this adaptation process is slow, uneven, and not suited for Twitter.
The actual trajectory is more volatile and less certain than the hype or panic suggests. Those who will do well from now on will be those who can endure this ambiguity, rather than those eager to grasp a prematurely concluded narrative.
Great operators always find a way out.
References
[1] Girard's Mimetic Desire Theory (https://www.iep.utm.edu/girard/)
[2] Fortune Magazine: Why SaaS Stocks Are Irrationally Declining Like During the DeepSeek Panic
[3] CNBC: The Impact of AI Tools on SaaS Software Stocks
[4] CNBC: Jensen Huang Calls AI Replacing Software "The Most Illogical Thing"
[5] Yahoo Finance: $285 Billion Evaporated from U.S. Software Sector Due to Anthropic Impact
[6] Yahoo Finance: Analysis of IGV ETF Trends
[7] Axios: Hedge Funds Heavily Shorting the Software Industry
[8] Benzinga: Misinterpretations in the Software Sector Crash
[9] Salesforce Investor Relations: Record Q3 Earnings Driven by Agentforce
[10] Futurum Group: ServiceNow Q4 Earnings and AI Platform Momentum
[11] Fortune Magazine: The AI Paradox and Irrational Analysis
[12] CNBC: Software Stocks Enter Bear Market, Significant Declines for Companies Like ServiceNow
[13] StockAnalysis: Datadog Operational Statistics
[14] Meritech Capital: Review of the 2022 SaaS Crash
[15] CNBC: Nvidia Plummets Due to DeepSeek Concerns
[16] Fortune Magazine: Deutsche Bank Discusses Software Stock Bubble and Internet Era Comparisons
[17] Deutsche Bank Jim Reid Analysis Report
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
