Sổ tay Đầu tư 2026: Tại sao một nhà phân tích cảnh báo về bong bóng tiền điện tử trong khi đặt cược lớn vào Google

Qiao Wang’s latest market assessment cuts through the optimism flooding financial markets, delivering a stark warning about inflated valuations even as she reveals a calculated 50/50 split between stocks and Bitcoin in her portfolio. The most striking element of her position: 40% of her total assets sit in cash, a defensive posture rooted in deep skepticism about unsustainable sentiment and the brewing crypto bubble that few seem willing to acknowledge.

In a January 2026 podcast appearance on Empire, Wang articulated a nuanced thesis that challenges mainstream enthusiasm. While the broader market sentiment has shifted decidedly bullish, with retail investors celebrating rapid gains on social media, her data-driven analysis suggests caution is warranted. “The market valuations are at historical highs,” she notes, echoing concerns also expressed by legends like Drucker, who recently called this “the most difficult moment” in his 50-year investing career to predict market movements.

Market Sentiment & the Risk of Excessive Optimism

Wang’s fear stems from what she describes as “overly optimistic market sentiment” colliding with stretched valuations. Though some argue today’s companies command premium multiples due to superior profitability and competitive moats, combined with the tailwinds of accommodative monetary policy, the fundamental issue remains unresolved: assets are priced at levels that leave little room for disappointment.

Her response reveals a sophisticated risk-management philosophy. Rather than timing the broader market—a task she acknowledges even legends can’t master reliably—Wang focuses intently on individual stock selection. This approach has historical precedent. During the dot-com crash of 2000, while tech stocks collapsed, small-cap and value equities delivered 10% average annual returns. “This is the best time for individual stock pickers,” she argues, suggesting that in periods of excessive concentration in certain assets or sectors, disciplined fundamental analysis can unearth diamonds while others panic.

The implications for crypto are particularly stark. Wang’s portfolio contains a modest allocation to Bitcoin and a minimal exposure to altcoins (less than 1%), reflecting her conviction that the crypto bubble presents more risk than opportunity in the near term. Yet she stops short of dismissing the entire asset class, maintaining that opportunities always exist—just not across the board. The lesson: asset quality varies dramatically, requiring granular analysis rather than category-level judgments.

Strategic Portfolio Construction: Google, Adobe, and the Art of Selective Conviction

Wang’s largest single investment is Google, a position born from personal usage patterns and rigorous fundamental analysis. After auditing her iPhone activity six months prior, she discovered her three most-used applications—Chrome, YouTube, and Gemini—were all Google products. This personal insight prompted deeper research that ultimately cemented her conviction.

The breakthrough: more than half of Google’s search revenue derives from shopping ads, a function that large language models like ChatGPT cannot currently replicate convincingly. Combined with proprietary assets like GCP (Google Cloud Platform), TPU (Tensor Processing Units), and YouTube’s vast video data library, Google’s competitive moats appear virtually unassailable in the near term. This analysis directly contradicts bearish narratives focused exclusively on search threat from AI.

Her other major holdings paint an interesting picture of selective value hunting. Tencent earns praise for “solid fundamentals” despite flying below the investment community’s radar. Amazon attracts her bullish stance primarily due to its decade-long robotics bet—an initiative that has already stabilized headcount while growing the robotic workforce by 20-30% annually, meaningfully improving profit margins.

Most intriguingly, Wang identifies Adobe as a sleeping giant. Despite widespread concerns that generative AI will cannibalize its products, her analysis reveals a critical overlooked factor: Adobe’s enterprise-level integration advantages are near-unbreakable. Creative professionals store years of work in Adobe Cloud, have internalized Photoshop’s interface through muscle memory, and face astronomical switching costs. At a P/E multiple of just 12—historically depressed for a high-quality enterprise franchise—Adobe mirrors the undervaluation Google faced years earlier. “Adobe could be the Google of this year,” she suggests, pointing to a market perception gap likely to close as AI’s actual impact on creative workflows becomes clearer.

AI’s Transformative Impact: The Reshaping of Software Moats and Startup Economics

The digital transformation driven by artificial intelligence presents a paradox that Wang articulates with precision. While the essence of competitive advantage hasn’t fundamentally changed, software moats are indeed weakening rapidly. Early-stage startups face near-zero defensibility, yet large incumbents like Google, Microsoft, and Adobe retain fortress-like positions rooted in switching costs, integrated ecosystems, and proprietary data.

The most provocative insight concerns startups themselves. AI-powered productivity tools enable extraordinarily small teams—potentially just one or two people—to generate meaningful revenue streams. Wang knows serial entrepreneurs running subscription businesses generating $10 million in annual revenue solo, typically veterans from Meta or Uber who rejected corporate bureaucracy. By 2026, she predicts, “unicorn startups with only one or two people” may already exist, not yet at billion-dollar valuations but expanding rapidly nonetheless.

This reframing of startup economics stems directly from AI’s productivity multiplier. When Claude Opus 4.5 or similar tools enable a solo operator to accomplish what previously required entire teams, the capital requirements for business formation collapse. “We don’t need to hire anyone,” becomes the recurring theme among early-stage founders Wang encounters. The emergence of specialized AI-assisted tools—sales commission calculators, data dashboards, analysis platforms—eliminates the need for entire support functions, compressing team structures while maintaining output.

Gemini’s Massive Undervaluation & the Code Automation Inflection Point

Wang’s assessment of Google’s Gemini tool reveals a striking valuation disconnect. “Gemini is underestimated by at least two orders of magnitude,” she states, suggesting that a $2,000/month price tag would remain a bargain for the tool’s capabilities as research assistant, junior engineer, junior analyst, medical advisor, and legal consultant combined.

The underlying shift deserves attention. With Claude Opus 4.5 and similar models, the bottleneck has fundamentally shifted. “The code itself is no longer the bottleneck; the key lies in designing appropriate prompts,” Wang emphasizes. This distinction matters enormously. Early AI tools could accelerate coding productivity in standard scenarios; modern systems can complete complex tasks end-to-end when specifications are sufficiently clear and comprehensive.

Her own experience building a Warren Buffett-inspired investment analysis system illuminates this transition. The actual code proved straightforward; crafting the prompt framework to simulate Buffett and Charlie Munger’s investment logic required months of iterative refinement across six distinct steps, combining deep research models for fact-gathering with inference models for reasoning. The system now filters thousands of stock codes, attempting to identify opportunities these legendary investors might pursue. Despite impressive results, Wang acknowledges current AI still requires human judgment for final investment decisions—though she predicts that gap will close by late 2026.

The Competitive Moat Paradox: Enterprise Software’s Enduring Strength

Wang challenges the narrative that AI renders all competitive advantages obsolete. While software moats are weakening, enterprise-level factors remain formidable: high switching costs, integrated ecosystems, proprietary data advantages, and vendor lock-in dynamics. AWS, Azure, and Google Cloud’s infrastructure dominance persists not through technical superiority alone, but through customer entrenchment. Microsoft’s enterprise software productivity suite faces replication, yet enterprises rarely switch—these tools are too critical, and transition costs too astronomical.

The insight carries significant implications for investment strategy. Companies offering to new users or competing primarily on feature parity face existential risk from AI-native competitors. Yet companies with decade-old customer relationships, proprietary data, and switching costs measured in millions retain strategic defensibility. This distinction explains why Wang remains confident in Google’s future despite ChatGPT’s rise, and why Adobe—despite the generative AI threat narrative—retains genuine fortress characteristics. The companies most at risk are those offering generic functionality to users who can easily switch.

Reframing Work, Automation, and the Future Labor Landscape

Wang argues forcefully that everyone must learn to code—though she carefully redefines the term. “This doesn’t refer to traditional programming, but rather to using automation tools to optimize certain aspects of one’s life and work,” she clarifies. As workflows become increasingly personalized and specialized, third-party software providers won’t develop tailored solutions for every niche use case. Individual workers and small teams will necessarily learn to deploy natural language interfaces to configure systems addressing their specific needs.

She acknowledges the labor market implications with candor. These tools will create a divergence: “AI will make already efficient and capable people even more efficient and intelligent, while those who are inefficient may fall further behind.” The broader pattern mirrors the internet’s adoption—transformative technology concentrates advantage rather than democratizing it. Ultimate outcomes depend less on the tool’s capability than on users’ willingness to deploy it.

Health, Longevity, and the Fallacy of Extreme Optimization

Wang’s perspective on health reveals a pragmatic philosophy grounded in years of experimentation. After exploring extreme optimization—testing supplements, saunas, and biohacking protocols—she concluded that fundamentals dominate: “The most important factors are still three basic ones: diet, sleep, and exercise.” Eight hours of consistent sleep, reasonable nutritional choices, and regular movement trump specialized interventions.

This wisdom extends a subtle critique of the obsessive optimization mindset gaining traction in Silicon Valley. “If I try to optimize everything to the extreme, I experience a lot of stress,” she notes. Elevated cortisol from chronic stress negates longevity benefits from other optimizations—a meta-pattern few wealthy investors fully internalize. The optimal approach often involves sustainable consistency over perfectionistic intensity.

The Path Forward: Selective Conviction in an Uncertain Market

Wang’s 2026 investment framework can be distilled to several core principles. First, avoid broad asset class judgments; instead, ruthlessly analyze individual opportunities. The crypto bubble may well arrive, yet Bitcoin may appreciate significantly. Valuation excess doesn’t preclude meaningful returns for select holdings. Second, identify companies with defensible competitive advantages strengthened rather than weakened by AI transformation—Google, Microsoft, and Adobe qualify, while feature-driven competitors face existential pressure.

Third, recognize that AI productivity gains will compress organizational structures and create new categories of possibility (solo-founder revenue businesses, AI-augmented venture capital, natural language automation across all industries). Fourth, price risk appropriately; holding 40% cash may feel costly during rallies but offers purchasing power when valuations normalize and opportunities emerge. Finally, apply consistent analytical rigor to market opportunity, recognizing that periods of excessive sentiment concentration create rare pockets of undervaluation for disciplined stock pickers.

As financial markets navigate 2026, the distinction between the broader asset bubble and specific pockets of genuine value has never been more pronounced. Wang’s playbook doesn’t promise to time market peaks or avoid crypto cycle downturns. It instead offers a framework for identifying companies and opportunities that merit conviction even in periods of excessive optimism, a reminder that disciplined fundamentals remain the most reliable compass when sentiment reaches extremes.