Khoo Says AI Trading Compresses Bear Markets, Widening Value Discounts

Adam Khoo

High-frequency algorithms compress bear markets from months to weeks by trading on short-term momentum rather than fundamentals, creating steeper pricing dislocations that patient investors can systematically exploit.

AI-assisted summary of Will AI Kill Value Investing? How to Beat the Algorithms in 2026. — original source. Check the original for full context.

Source published: 2026-08-14

Key ideas

Rather than making markets hyper-efficient, algorithmic trading—which now drives 60% to 70% of trading volume—has compressed 20% bear-market declines from a historical average of ~250 days down to less than 30 days, creating faster and deeper valuation discounts for long-term investors. (16:43)
Even heavily scrutinized mega-cap stocks continue to experience major mispricings, with intrinsic value scans showing discounts of roughly 38% on ServiceNow, 34% on Meta Platforms, and nearly 12% on Microsoft. (3:42)
Since the launch of ChatGPT on November 30, 2022, Adam Khoo's Value Momentum Investing approach has generated a 137% cumulative return against the S&P 500's 90% (and 25.66% YTD vs. 13.00%), demonstrating that systematic quality-value investing outperforms in the AI era. (5:28)
Relying on generic AI chatbots to pick stocks fails because LLMs synthesize historical market consensus—eliminating any differentiated edge—while hallucinating financial figures like EPS and valuation ratios up to 39% of the time. (22:47)

Detailed notes

## Background Published by Adam Khoo on YouTube on August 14, 2026, this video addresses the growing retail anxiety that artificial intelligence and algorithmic trading are rendering value investing obsolete. According to the show notes, Khoo sets out to debunk common myths, explaining why AI stock-picking fails, how high-frequency machines create faster valuation discounts, and how his Value Momentum Investing framework capitalizes on machine-driven volatility.

## TL;DR Adam Khoo counters the prevailing narrative that AI and high-frequency algorithms have eliminated market inefficiencies, demonstrating that mega-cap leaders continue to trade at significant discounts to intrinsic value. He argues that using general AI models to pick stocks guarantees mediocre consensus returns, while general chatbots frequently hallucinate critical financial data with unearned confidence. Because algorithms trade on microsecond timeframes without regard for business fundamentals, they dramatically compress market cycles and produce sharper, faster sell-offs. Investors can exploit these algorithmic overreactions by filtering strictly for the top 1% highest-quality businesses and systematically accumulating shares in tranches at strong technical support levels.

## Atmosphere & style The presentation carries the brisk, confident energy of a seasoned market veteran brushing aside modern techno-panic with historical perspective. Khoo speaks in a candid, engaging instructional style, blending technical charts and hard valuation data with sharp humor, relatable dating analogies, and cautionary parables about AI hallucinations. The delivery feels pragmatic and empowering, steering viewers away from paralysis and framing algorithmic volatility not as an existential threat, but as an exceptional profit opportunity.

## Bulletin - [16:43](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1003s) Rather than making markets hyper-efficient, algorithmic trading—which now drives 60% to 70% of trading volume—has compressed 20% bear-market declines from a historical average of ~250 days down to less than 30 days, creating faster and deeper valuation discounts for long-term investors. - [03:42](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=222s) Even heavily scrutinized mega-cap stocks continue to experience major mispricings, with intrinsic value scans showing discounts of roughly 38% on ServiceNow, 34% on Meta Platforms, and nearly 12% on Microsoft. - [05:28](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=328s) Since the launch of ChatGPT on November 30, 2022, Adam Khoo's Value Momentum Investing approach has generated a 137% cumulative return against the S&P 500's 90% (and 25.66% YTD vs. 13.00%), demonstrating that systematic quality-value investing outperforms in the AI era. - [22:47](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1367s) Relying on generic AI chatbots to pick stocks fails because LLMs synthesize historical market consensus—eliminating any differentiated edge—while hallucinating financial figures like EPS and valuation ratios up to 39% of the time.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=0s) Underperforming investors and fund managers frequently blame AI—citing instant fundamental screening, homogenized retail prompts, and microsecond institutional HFT execution—to rationalize poor returns and justify giving up on active investing. - [02:28](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=148s) Claims that new technology will permanently kill active investing recur every decade, echoing past panics over Bloomberg terminals in the 1980s, S&P 500 index funds in the 1990s, and quantitative arbitrage funds in the 2000s. - [03:42](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=222s) Substantial valuation gaps persist even across heavily covered trillion-dollar and mega-cap equities, including ServiceNow (~38% discount), Meta Platforms (~34% discount), Eli Lilly (12% discount), Microsoft (11.98% discount), and UnitedHealth Group (10% discount). - [05:28](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=328s) Real-world performance refutes the death of value investing: since ChatGPT launched on November 30, 2022, the Value Momentum Investing framework returned 137% versus 90% for the S&P 500, and 25.66% YTD through August 13 versus the benchmark's 13.00%. - [08:49](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=529s) Asking an LLM what stock to buy guarantees mediocrity because AI models operate on historical data and mathematical probability, synthesizing existing crowd sentiment and analyst consensus rather than generating a differentiated edge. - [10:12](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=612s) While intraday retail traders cannot compete with institutional supercomputers on microsecond latency, long-term investors play a multi-year game where mechanical liquidity churning actually widens short-term pricing inefficiencies. - [10:38](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=638s) High-frequency trading robots execute 60% to 70% of all market volume with complete disregard for intrinsic business value, amplifying systemic volatility and driving steeper sell-offs and faster rebounds. - [11:26](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=686s) The 'quality value investing' philosophy prioritizes business caliber over cheap multiples, restricting investments to the top 1% of companies featuring wide economic moats, consistent revenue and cash flow growth, and ROIC and ROE exceeding 12% to 15%. - [14:01](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=841s) Because short-term emotions and algorithmic momentum cause share prices to swing wildly around a rising intrinsic value curve, investors should accumulate undervalued stocks in structured tranches at technical support levels rather than attempting to time exact bottoms. - [16:43](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1003s) Automated trading has compressed multi-year market cycles into weeks: while 20% bear-market declines took ~250 days in the pre-algo era and just over 100 days in 2022, both the COVID crash and the 2025 tariff sell-off dropped 20% in under 30 days. - [20:20](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1220s) Fundamental valuation modeling places Microsoft's intrinsic value near $570 and Meta Platforms' near $900 (~34% discount), with big tech's large debt and capex figures easily supported by fortress balance sheets and massive operating cash flows. - [22:14](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1334s) Meta Platforms' chart exhibits a rhythmic wave structure currently resting on a strong technical support floor, presenting a compelling risk-to-reward entry point ahead of its next upward cycle. - [22:47](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1367s) Generic consumer chatbots like ChatGPT, Claude, Gemini, and Grok hallucinate financial data—quoting stale prices, wrong EPS, and invented valuation ratios—up to 39% of the time with complete confidence, requiring investors to use specialized, data-verified financial tools for research.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=0s) The Panic Over AI and Value Investing

A narrative has taken hold across financial circles that artificial intelligence is making value investing obsolete. As retail traders adopt AI-driven stock pickers, investors lean on generative tools for research, and major institutions deploy machine-learning models into high-frequency trading (HFT), many market participants feel outmatched. The prevailing fear is that markets have become hyper-efficient, squeezing out every edge and making it impossible to consistently beat benchmark indices.

```mermaid flowchart TD accTitle: The AI Efficiency Argument vs Investor Reality accDescr: Flowchart illustrating the common fear that AI tools eliminate all market edges versus the resulting investor despair.

subgraph Fear["The Market Narrative"] A["Retail AI Stock Picking"] --> D["Hyper-Efficient Market"] B["Instant Fundamental Research"] --> D C["Algorithmic HFT Execution"] --> D end

subgraph Outcome["The Flawed Conclusion"] D --> E["No Undervalued Stocks Left"] E --> F["Edge Eroded: Give up & Buy the Index"] end ```

Underperforming investors and struggling institutional fund managers frequently rationalize their poor returns by blaming AI across three main arguments: - **Instantaneous Fundamental Analysis:** In previous eras, uncovering deeply undervalued businesses required spending hours digging through financial reports and manually running discounted cash flow models. Today, automated tools compute intrinsic values within seconds. Skeptics argue that because screening occurs instantaneously, undervaluation disappears before humans can react. - **Homogenized Stock Selection:** Millions of retail investors use identical AI prompts asking, *"What stock should I buy?"* Critics claim this widespread reliance on uniform models funnels capital into identical tickers, eliminating the distinct information edge required to outperform the S&P 500 or Nasdaq 100. - **Institutional Algorithmic Execution:** Supercomputers powered by complex neural networks execute orders within microseconds. Human discretionary traders compare facing these machines to playing chess against a grandmaster supercomputer, assuming automated market makers capture every penny of edge.

---

[02:28](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=148s) A Familiar Excuse Across Market Decades

The panic that modern technology will permanently destroy active market participation is entirely unfounded. Looking back across financial history reveals that market participants advance nearly identical doomsday arguments roughly once every decade:

Just as previous technological shifts failed to eliminate market mispricings, modern algorithmic dominance has not broken value investing. Blaming AI is simply the latest excuse for flawed strategies and poor execution.

---

[03:42](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=222s) Evidence: Large-Cap Mispricings Persist

The claim that instant AI valuation prevents high-quality companies from becoming deeply undervalued collapses under scrutiny. Even within the highly scrutinized mega-cap universe, substantial valuation gaps appear continuously. A quick scan of tools like the AI-powered Stock Oracle Value Radar shows quality assets trading well beneath fair value:

- **Microsoft (MSFT):** Despite heavy research coverage from institutional analysts, trading at an intrinsic valuation discount of **11.98%**. - **Meta Platforms (META):** Trading at an estimated discount of **34%**. - **Eli Lilly (LLY):** Undervalued by **12%** (though personal investment preference may avoid pharmaceutical developers). - **ServiceNow (NOW):** Trading at roughly a **38%** discount to fair value. - **UnitedHealth Group (UNH):** Undervalued by **10%** (having traded at near a 50% discount when accumulated previously). - **Recent Opportunities:** Mega-cap tech names like Alphabet (Google), Nvidia, and Netflix offered significant double-digit margins of safety only weeks and months prior.

If automated algorithms instantly priced equities at perfection, wide valuation discounts across trillion-dollar companies could not exist.

---

[05:28](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=328s) Performance Data and Real-World Outperformance

Real-world returns demonstrate that a systematic framework outperforms market benchmarks even as algorithmic adoption accelerates. Since the launch of ChatGPT on November 30, 2022—the turning point marking modern consumer and institutional AI integration—the Value Momentum Investing approach has widened its outperformance over the broader indices:

- **Year-to-Date Return:** Generating **25.66%** through August 13 against the S&P 500's **13.00%**. - **Post-AI Launch Period (Nov 30, 2022 to Present):** Generating a cumulative return of **137%**, compared to the S&P 500's **90%**. - **Account Transfer Detail:** A severe balance drawdown shown on portfolio equity charts during February reflects an operational transfer rather than a trading loss; 99% of funds were moved from an individual personal account into a corporate entity to safeguard against prospective US estate taxes on non-resident accounts.

Student case studies demonstrate that disciplined execution consistently captures market alpha: - One long-term student joining August 26, 2024, transitioned away from haphazard stock picks toward a rule-based system, securing **$445,000 in net profit** with a **37% gain**. - A student enrolling in the Whale Investor course in February 2026 generated a **42% year-to-date return** focusing strictly on top-percentile companies.

---

[08:49](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=529s) Why Using AI to Pick Stocks Guarantees Mediocrity

Neither successful systematic investors nor their students rely on AI prompts to select portfolio holdings. Asking an LLM, *"What stock should I buy?"* is a guaranteed route to underperformance. Large language models excel at synthesizing information, parsing earnings transcripts, and organizing financial data, but fail fundamentally at alpha generation.

```mermaid flowchart LR accTitle: The Consensus Loop of AI Stock Picking accDescr: Flowchart showing how AI models synthesize past consensus data into crowded trades with zero alpha.

A["Historical Data & Consensus Reports"] --> B["AI Processing Engine"] B --> C["Consolidated Crowd Sentiment"] C --> D["Zero Differentiated Edge"] ```

AI operates entirely on historical datasets and mathematical probability, effectively synthesizing market consensus. When prompted for investment ideas, the algorithm compiles whatever the market already accepts: popular sentiment, widely published analyst ratings, and established narratives.

Buying an AI's chosen equities simply executes consensus crowd psychology. Because outperforming a benchmark index requires holding a differentiated view from the crowd, relying on an engine trained on the crowd cannot work: "When everyone thinks they've got an edge, no one has an edge... AI *is* the market."

---

[10:12](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=612s) Time Horizons: Retail Investors vs. Machine Speed

The argument that algorithmic trading drains market profits applies selectively depending on an investor's operational timeframe: - **Intraday Traders:** Attempting to scalp micro-movements on short timeframes puts individual retail day traders in direct competition with collocated institutional supercomputers executing orders in microseconds. Competing on raw latency is mathematically unsustainable. - **Long-Term Value Momentum Investors:** High-frequency algorithms focus purely on millisecond price discrepancies rather than multi-year corporate cash flows. This mechanical, ultra-fast liquidity churning does not eliminate value-oriented opportunities—it exacerbates short-term market inefficiencies, creating larger pricing dislocations for patients investors to exploit.

[10:38](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=638s) The Algorithmic Reality: High-Frequency Machines and Market Volatility

Currently, between 60% and 70% of all buying and selling transactions across the stock market are executed not by human beings, but by high-frequency trading robots. These algorithms operate with complete disregard for intrinsic value or underlying business fundamentals. Instead, their decisions are governed strictly by short-term price action, technical signals, and execution speeds.

This dominance of algorithmic trading fundamentally alters market dynamics by drastically ramping up systemic volatility. Rather than creating smooth, efficient price curves, machine trading exaggerates both directions: sell-offs become steeper and swifter, while subsequent rebounds occur with violent speed. Far from harming long-term investors, this heightened volatility creates an unprecedented advantage for disciplined practitioners. By driving share prices away from business realities at machine speed, algorithms create regular, deep pricing dislocations that investors can systematically exploit.

[11:26](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=686s) The Quality Value Investing Philosophy and the Top 1% Filter

Adam Khoo clarifies that while he is widely categorized as a value investor, his actual framework is best described as "quality value investing." Under this approach, valuation is secondary to business caliber; an investor must never buy a mediocre company simply because the stock appears cheap on a multiples basis. The core objective is to identify and own only the top 1% highest-quality companies in the entire market, which possess durable competitive advantages and predictable financial performance.

```mermaid flowchart TD accTitle: Quality Value Filter accDescr: Criteria used to screen for top 1 percent quality businesses before considering valuation. subgraph S1["Quality Screening Criteria"] A["Wide Economic Moat (High barrier to entry)"] B["Consistent Trajectory (Revenue, Op Income & OCF)"] C["High Capital Efficiency (ROIC > 12-15%, ROE > 12-15%)"] end subgraph S2["Valuation Layer"] D["Stock Oracle AI Valuation (Intrinsic Value Calculation)"] E["Margin of Safety Entry (Accumulate in Tranches)"] end S1 --> S2 ```

To demonstrate how an investor can evaluate these criteria in seconds, Khoo references Stock Oracle, an AI-enabled fundamental analysis tool. Taking Microsoft as an example, the tool quickly assesses predictable cash flow, profitability, growth rate, economic moat, and balance sheet resilience:

- **Wide Economic Moat:** The company must have virtually impenetrable defenses against disruption or intense market competition. - **Consistent Top- and Bottom-Line Expansion:** A multi-year history of uninterrupted upward trajectory across revenue, operating income, and net operating cash flow. - **Superior Return Metrics:** High return on invested capital (ROIC) and return on equity (ROE) well above the baseline hurdle of 12% to 15%. Microsoft, for instance, registers an ROIC of 26% and an ROE of 34%. - **Selectivity:** Less than 1% of all publicly traded equities satisfy these strict fundamental benchmarks.

[14:01](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=841s) Intrinsic Value vs. Sentiment Waves: How to Accumulate in Tranches

The primary reason to restrict investments to the top 1% is the mathematical conviction that their intrinsic value must inevitably rise over time as the companies generate expanding pools of operational cash. However, share price trajectories never climb in a smooth, straight line. In the short term, market pricing is dictated by emotions, headline news, macro sentiment, and deliberate price manipulation driven by high-frequency algorithms.

When positive headlines and algorithmic momentum push a stock up, "the stock price can go whoa!" surging well above its underlying fair value. Conversely, when panic, bad news, or technical sell orders hit, the price overshoots downward, sinking deeply below intrinsic value. These downward deviations represent the investor's designated "buy zone."

Khoo emphasizes that attempting to pinpoint the absolute market bottom is a fool's errand. Instead of trying to catch falling knives at exact lows, investors should initiate accumulation once a stock trades at a sufficient margin of safety—such as 20% below intrinsic value—and execute buys in structured tranches at technical support levels:

- Rather than going all-in at once, allocate capital in tranches (e.g., buying a quarter position at an initial support zone). - If algorithmic selling drags the price lower, purchase a second quarter. - When price action stabilizes and starts to rebound, add the next quarter. - This tranche approach achieves a favorable average cost near cyclical lows without requiring market-timing perfection, allowing capital to compound smoothly as price eventually tracks the rising intrinsic value curve upward.

[16:43](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1003s) Compressing the Cycles: Why Faster Crashes Benefit Modern Value Investors

In the pre-AI era, market cycles unfolded at a glacial pace. Before the rise of algorithmic trading networks, it often took many months or years for stock prices to slowly grind below intrinsic value, and subsequent recoveries to fair value took years to materialize. In contrast, modern automated markets compress multi-year cycles into a matter of weeks.

Khoo points to historical market data illustrating how machine execution has magnified volatility spikes and truncated bear markets:

| Era / Market Cycle Event | Historical Dynamics & Speed to 20% Bear Market Drop | |:--- |:--- | | **Pre-Algo Era (1997–2000s)** | Historically took an average of **~250 days** (nearly three quarters of a year) for the market to drop 20% into bear market territory. Major crisis spikes in the VIX generally crested around 50. | | **Algo Era (2018–2022)** | In the 2022 bear market, the 20% decline was compressed down to **just over 100 days**. Major VIX volatility spikes reached much higher peaks during systemic shocks. | | **Modern AI/Algo Era (COVID & 2025 Tariff Crash)** | In the 2025 tariff sell-off, the 20% drop occurred in **less than 30 days**; during the COVID crash, the 20% drop materialized in **less than a month**. |

Rather than fearing these violent drop-offs, Khoo views them as "fantastic" for disciplined investors. Machine-induced liquidations trigger sharp V-shaped market recoveries. Investors who keep their emotions in check can now buy into top-tier businesses at 30% to 40% discounts far more frequently than previous generations of investors ever could.

[20:20](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1220s) Real-World Case Studies: Analyzing Microsoft and Meta Platforms

Khoo walks through live chart and valuation breakdowns of two mega-cap leaders to illustrate the difference between buying at deep bottoms versus accumulating mid-rebound:

- **Microsoft ($MSFT):** The stock demonstrates a clear recurring rhythm of alternating waves up and waves down. A prime accumulation window occurred when the stock hit support and printed a clean double-bottom pattern. With an estimated intrinsic value of approximately $570, Microsoft remains undervalued, though Khoo notes that buying after it has already completed half of its upward wave is noticeably less advantageous than catching it directly at the support floor. - **Meta Platforms ($META):** Meta presents an exceptionally compelling risk-to-reward ratio. According to Stock Oracle's algorithmic valuation and discounted cash flow modeling, Meta's intrinsic value stands at roughly $900 per share, leaving the stock trading at a deep discount of approximately 34%. - **Addressing Debt Concerns:** Khoo addresses investor anxiety over big tech's soaring debt loads and massive capital expenditure budgets. While headline debt figures reach tens of billions of dollars, they remain entirely manageable when measured against the companies' fortress balance sheets, massive cash and cash equivalents, and relentless operational cash flow generation. Meta scores at the top of the spectrum across predictability, profitability, economic moat strength, and overall balance sheet health.

[22:14](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1334s) Meta at Key Support and Market Patience

Analyzing Meta Platforms, Adam notes that he personally already holds a full position and is not adding more shares, but emphasizes that if an investor does not yet have enough Meta in their portfolio, the current level presents an attractive entry point: "if I didn't have enough Meta, would I add more here? I wouldn't mind." Looking at Meta's price chart reveals a steady, rhythmic wave structure: "wave up, wave down, wave up, wave down..." repeatedly bouncing off a well-defined price floor.

The stock is resting right at a very strong level of support. Using his familiar dating analogy, Adam remarks that "the girl hasn't run away yet"—the stock is still patiently waiting at support rather than breaking down, setting the stage for what should eventually be a substantial move to the upside. The market continues to offer abundant opportunities for investors who stay patient and concentrate their capital on great companies.

[22:47](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1367s) The Danger of AI Hallucinations in Financial Research

While AI can be an exceptional tool for research when deployed correctly, Adam reiterates that it is outright terrible at actually picking stocks. Using generic consumer chatbots like ChatGPT, Gemini, or Grok for financial analysis carries massive risk because roughly 30% or more of their output consists of pure hallucinations—data the model simply invents.

When questioned about their own reliability, modern frontier models will openly admit to this flaw. Adam shares that he directly asked Claude how much data in its financial prompt responses is hallucinated. The model admitted that chatbots state wrong earnings per share (EPS), quote stale stock prices, and deliver completely invented valuation ratios—all with total, unshakeable confidence.

```mermaid flowchart TD accTitle: Generic LLM Financial Research Risks accDescr: Diagram illustrating how general AI chatbots hallucinate financial data leading to bad investment decisions.

subgraph Prompting ["User Inquiry"] A["Retail Investor"] -->|"Prompts general LLM (ChatGPT / Claude / Gemini)"| B["Standard Chatbot Interface"] end

subgraph ModelOutput ["Model Processing & Flaws"] B --> C{"Up to 39% Hallucination Rate"} C --> D["Wrong Earnings Per Share (EPS)"] C --> E["Stale / Lagging Stock Prices"] C --> F["Completely Invented Valuation Ratios"] end

subgraph Result ["High-Confidence Failure"] D & E & F --> G["Delivered with 'Total Confidence'"] G --> H["Severe Capital Loss / Bad Trades"] end ```

Industry findings show that chatbots produce incorrect financial facts up to 39% of the time. Adam asks the audience incredulously how anyone could possibly stake their hard-earned money on an engine that feeds them made-up numbers almost four out of ten times. He compares this blind trust to a dark cartoon strip:

> A man holds up a wild mushroom and asks an AI chatbot, *"Is this mushroom edible?"* > > The AI responds with supreme confidence: *"Yes."* > > The man eats it and instantly drops dead. The AI immediately follows up: *"You're right. It's a poisonous mushroom. Would you like to learn more about poisonous mushrooms?"* > > The punchline: *"Well, no. I'm dead."*

To protect against this, specialized AI-enabled tools like Stock Oracle are built to ground algorithmic research in verified data feeds so investors do not rely on hallucinated figures.

- Standard chatbots invent financial figures, misstate EPS, and cite stale market prices up to 39% of the time. - Generic LLMs deliver false facts with the exact same authoritative tone as factual ones. - Purpose-built financial platforms like Stock Oracle are necessary to constrain AI to accurate research.

[24:19](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1459s) Channel Outro and Wealth Academy Resources

Wrapping up the video, Adam urges viewers to hit the subscribe button and click the notification bell to get immediate alerts whenever a new market breakdown is released. For those looking to build consistent income streams and master the craft of trading and investing across global financial markets, he points viewers to the training programs hosted at Piranha Profits (`piranhaprofits.com`).

He also invites traders to explore the live interactive mentorship curriculum through Wealth Academy Live (`wealthacademylive.com`), where students learn hands-on strategies live online. Signing off with his trademark blessing to the trading community, Adam leaves the audience with: "This is Adam Khoo, and may the markets be with you."

## Notable quotes > [00:00](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=0s) What stock should I buy? — Adam Khoo

> [08:49](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=529s) When everyone thinks they've got an edge, no one has an edge... AI *is* the market. — Adam Khoo

> [14:01](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=841s) the stock price can go whoa! — Adam Khoo

> [22:14](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1334s) if I didn't have enough Meta, would I add more here? I wouldn't mind. — Adam Khoo

> [22:14](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1334s) wave up, wave down, wave up, wave down... — Adam Khoo

> [22:14](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1334s) the girl hasn't run away yet — Adam Khoo

> [22:47](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1367s) Is this mushroom edible? — Adam Khoo

> [22:47](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1367s) You're right. It's a poisonous mushroom. Would you like to learn more about poisonous mushrooms?

> [22:47](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1367s) Well, no. I'm dead.

> [24:19](https://www.youtube.com/watch?v=QzqQwF9hCNY&t=1459s) This is Adam Khoo, and may the markets be with you. — Adam Khoo

## Mentioned Adam Khoo, Alphabet, Bloomberg Terminal, ChatGPT, Claude, Eli Lilly, Gemini, Grok, Meta Platforms, Microsoft, Nasdaq 100, Netflix, Nvidia, Piranha Profits, S&P 500, ServiceNow, Stock Oracle, UnitedHealth Group, Wealth Academy, Whale Investor

## Video description Will AI kill value investing? Not even close! but it IS changing everything.

62% of retail investors now use AI to pick stocks. Algorithms run 60-70% of all trading volume. And a growing fear is spreading: that machines will make markets so efficient, there will never be an undervalued stock again.

In this video, I will show you the facts

I'll show you: • Why AI stock-picking fails, when everyone has the same edge, nobody has an edge • How AI is trained on consensus, so it literally becomes the market it's trying to beat • Why the algorithms play a completely different game on a different time frame than you do • How machine-speed trading is creating bigger, faster discounts than we've ever seen • The right way to use A... as your analyst, never your fund manager • The exact Value Momentum Investing method my students use to profit from all of it

======================== Stock Investing & Trading insights by Adam Khoo will show you profitable trading and investment opportunities in today's stock markets.

These are essential strategies for stock traders and investors who want to improve their investment and trading performance.

Adam Khoo is a professional stock investor and options trader and the best-selling author of 16 books including 'Winning the Game of Stocks" and "Profit from the Panic". Thousands of students have profited from his sharp investment insights into the world of stock market investing and trading. ===================== Sign up for our courses/mentorship programs 📗 Value Momentum Investing: http://bit.ly/4mqrEVP 📙 Options Waverider™: http://bit.ly/4c5xyrR 📘 Price Action Manipulation™ (Trading): http://bit.ly/48FUAmQ 📕 Income Investing™ (Dividends): http://bit.ly/3Ow8CRj

Grab our Playbook Subscriptions Ultimate Investors Playbook™: http://bit.ly/3QiN6zT Options Tactics Playbook™: http://bit.ly/4tCQFPL PAM™ Trader Playbook: http://bit.ly/4mmYnv0 ===================== 💻 Join Our Wealth Academy Live Seminars https://www.wealthacademyglobal.com ===================== 🧑‍🏫 Visit Adam Khoo Learning Technologies Group https://www.akltg.com ===================== Follow Me On Social Media 🌟 Facebook: https://facebook.com/adamkhoosuccess 🌟 Instagram: https://www.instagram.com/adamkhoosuccess/ ===================== Follow Piranha Profits On Social Media 🐟 Facebook: https://www.facebook.com/piranhaprofits 🐟 Instagram: https://www.instagram.com/piranhaprofits 🐟 Telegram: https://t.me/piranhaprofits 🐟 Discord: https://discord.gg/piranhaprofits ===================== 👉 If you've found this video helpful, please LIKE, SHARE, & COMMENT 👉 SUBSCRIBE to my Channel for more Market Analyses & Investing Strategies 👉 Click on the NOTIFICATION BELL to get notified whenever I upload a new video ===================== 💸 MAY THE MARKETS BE WITH YOU! 💸

[Source](https://www.youtube.com/watch?v=QzqQwF9hCNY)

中文简报

Khoo称AI交易压缩熊市周期，拉大价值折价空间高频算法基于短期动量而非基本面进行交易，将熊市周期从数月压缩至数周，从而造成更剧烈的价格错位，有耐心的投资者可对此进行系统性利用。 Adam Khoo反驳了当下关于AI和高频算法已消除市场无效性的普遍论调，指出超大盘龙头股的交易价格仍相对其内在价值存在显著折价。他认为，利用通用AI模型选股注定只能获得平庸的共识回报，而通用聊天机器人往往会以毫无依据的自信捏造关键财务数据。由于算法以微秒级的时间维度交易且完全不考虑商业基本面，它们极大地压缩了市场周期，并导致更剧烈、更迅猛的抛售。投资者可以通过严格筛选排名前1%的最高质量企业，并在强技术支撑位分批系统性建仓，从而利用这些算法驱动的过度反应获利。算法交易目前已占据60%至70%的交易量，但它非但没有让市场变得极度高效，反而将幅度达20%的熊市下跌周期从历史上平均约250天压缩至不到30天，为长期投资者创造了更迅速且更深度的估值折价。 (16:43) 即便是受到密切审视的超大盘股也持续出现严重定价错误，内在价值筛查显示，ServiceNow的折价幅度约为38%，Meta Platforms约为34%，微软接近12%。 (3:42) 自2022年11月30日ChatGPT推出以来，Adam Khoo的价值动量投资法已取得137%的累计回报，而标普500指数同期为90%（年初至今为25.66%，标普为13.00%），证明系统化的优质价值投资在AI时代依然能跑赢大盘。 (5:28) 依靠通用AI聊天机器人选股行不通，因为大语言模型综合的是历史市场共识，从而消除了任何差异化优势，同时它们在捏造每股收益（EPS）和估值比率等财务数据方面的概率高达39%。 (22:47)

Reading summary

Adam Khoo argues that artificial intelligence and algorithmic trading do not destroy value investing, but instead create faster, wider pricing dislocations that disciplined investors can systematically exploit. While generic AI models merely synthesize market consensus and frequently hallucinate financial figures, applying fundamental quality screens allows long-term investors to consistently outperform automated markets. Technological Panic and Persistent Large-Cap Mispricings (0:00) Financial circles periodically advance narratives that technological breakthroughs eliminate active market edges, citing computerized charting in the 1980s, passive indexing in the 1990s, quantitative funds in the 2000s, and generative AI today. Critics claim that instant automated screening and algorithmic execution squeeze out all undervalued stocks. However, Khoo contends this is merely an excuse for underperformance, pointing out that significant valuation discounts persist even among heavily analyzed mega-cap companies. Using the Stock Oracle tool, Khoo demonstrates substantial gaps between market prices and estimated intrinsic values: Microsoft traded at an 11.98% discount, Meta Platforms at roughly 34%, ServiceNow at 38%, Eli Lilly at 12%, and UnitedHealth Group at 10%. Furthermore, Khoo reports that his Value Momentum Investing framework achieved a 137% cumulative return from the launch of ChatGPT on November 30, 2022, through mid-August, compared to the S&P 500's 90%, evidencing that algorithmic dominance has not erased active outperformance. Why AI Stock Picking Generates Mediocrity (8:49) Relying on artificial intelligence prompts to pick stocks produces poor returns because large language models inherently synthesize historical datasets, popular sentiment, and prevailing consensus ratings. Because beating market benchmarks requires a differentiated viewpoint, relying on models that reflect aggregate crowd psychology yields zero informational edge. Furthermore, generic consumer chatbots frequently generate inaccurate financial data, with error and hallucination rates cited up to 39% across metrics such as earnings per share, valuation ratios, and stock prices, delivered with misleading confidence. Algorithmic Speed and Compressed Market Cycles (10:12) While retail day traders face an impossible mathematical disadvantage competing against institutional high-frequency trading (HFT) supercomputers on microsecond latency, long-term investors face no such barrier. Algorithms account for 60% to 70% of total market transactions and execute purely on technical momentum and speed rather than intrinsic business value. Consequently, machine trading magnifies market volatility, steepening sell-offs and accelerating market rebounds. This algorithmic execution compresses market cycles: dropping 20% into bear market territory required roughly 250 days during the 1997–2000s era, compared to just over 100 days in 2022, and under 30 days during the COVID and 2025 tariff declines. Rather than harming investors, Khoo asserts that these accelerated drawdowns create sharper V-shaped recoveries and more frequent opportunities to accumulate top-tier businesses at 30% to 40% margins of safety. Quality Value Screening and Tranche Accumulation (11:26) Khoo outlines a "quality value investing" strategy that prioritizes business fundamentals over low valuation multiples, filtering strictly for the top 1% of public companies. Qualifying businesses must possess wide economic moats, uninterrupted multi-year growth across revenue, operating income, and operating cash flow, and sustained capital efficiency metrics with ROIC and ROE exceeding 12% to 15%—such as Microsoft's 26% ROIC and 34% ROE. Because short-term price movements are driven by emotion and algorithmic momentum rather than steady intrinsic value growth, Khoo advises against trying to time exact bottoms. Investors should instead build positions in structured tranches—such as quarter allocations at technical support levels—once prices fall 20% or more below intrinsic value. Case studies in Microsoft and Meta Platforms illustrate this approach, with Meta identified at strong chart support trading near $594 against a modeled intrinsic value of roughly $900, backed by solid cash flow despite broader market anxiety over enterprise debt loads and capital expenditures. The presentation concludes that disciplined value investing thrives on algorithmic volatility rather than succumbing to it, provided investors avoid using generic AI prompts for trade selection. However, the framework relies heavily on proprietary valuation benchmarks and requires strict emotional discipline to execute tranche purchases during steep market sell-offs.

Source quotes

[0:00] Adam Khoo: What stock should I buy? [8:49] Adam Khoo: When everyone thinks they've got an edge, no one has an edge... AI *is* the market. [14:01] Adam Khoo: the stock price can go whoa! [22:14] Adam Khoo: if I didn't have enough Meta, would I add more here? I wouldn't mind. [22:14] Adam Khoo: wave up, wave down, wave up, wave down... [22:14] Adam Khoo: the girl hasn't run away yet [22:47] Adam Khoo: Is this mushroom edible? [22:47] You're right. It's a poisonous mushroom. Would you like to learn more about poisonous mushrooms? [22:47] Well, no. I'm dead. [24:19] Adam Khoo: This is Adam Khoo, and may the markets be with you.

Original transcript

[0:00] There are some people who believe that [0:01] AI is killing value investing because as [0:05] more and more retail investors use AI to [0:08] pick stocks, use AI to do stock [0:09] research, and more institutions are [0:11] using AI in high-frequency trading, it's [0:14] becoming very difficult or almost [0:16] impossible to profit and to beat the [0:18] market. Now, is this true or is AI [0:22] actually making it easier for people [0:24] like me to beat the market? Let me show [0:26] you the facts in this video. [0:37] There are many people who are losing [0:39] money in the stock market or they're not [0:41] doing as well, even some professional [0:43] fund managers, and guess what? They are [0:45] blaming AI. And there are three reasons [0:48] why they blame AI. Number one, they say [0:50] because of AI, it is impossible to find [0:52] good companies that are undervalued. [0:54] Why? See, in the old days, uh you know, [0:57] you had to spend hours and hours to [0:59] research a company and to calculate the [1:01] intrinsic value. And so, because of [1:04] that, you could find great companies [1:06] that were undervalued that people [1:07] couldn't find. But now with AI, anyone [1:10] can research a stock to you can value a [1:12] stock in seconds with the right tools. [1:15] So, if everyone can analyze stocks so [1:17] fast, then there are no more undervalued [1:20] stocks because everyone has bought it, [1:22] right? So, that's the first argument. [1:24] Second argument they make is because [1:27] uh many retail investors now use AI to [1:29] pick stocks. They say, "Okay, tell me [1:31] what stock to buy." If everyone knows [1:33] what stocks to buy, everyone buys the [1:34] same stocks, it erodes the advantage of [1:37] any kind of stock picking. Because of [1:39] that, there's no way to beat the market, [1:41] to beat the S&P 500 or the Nasdaq 100. [1:44] The third argument is that because more [1:46] and more institutions are using [1:48] high-frequency [1:50] trading with [1:52] AI and with machine learning, these [1:54] supercomputers, they buy and sell in [1:58] and it's impossible for us to compete [2:00] with with these super computers. It's [2:02] kind of like playing chess with a super [2:04] computer. You cannot win them. And [2:07] that's why most traders are finding that [2:08] they can't make any money because any [2:10] profit we made has been taken by the [2:12] super computers and they are left with [2:14] nothing. Is this true? Is it true that [2:17] everyone is using AI so we can't beat [2:19] the market anymore? We can't do value [2:20] investing. We give up. WE JUST BUY the [2:22] index. No, it is not true and I'm going [2:25] to prove to you why it is not true. [2:28] First of all, understand that this [2:29] argument that this new technology [2:32] changes the market is not something new. [2:36] I remember that back in the 1980s when [2:38] they first came up with the Bloomberg [2:40] terminal where people said that now you [2:41] can analyze markets, analyze stocks with [2:44] computers with the Bloomberg terminal. [2:45] They said, "Oh, that would change the [2:46] market with this. No one can make money [2:47] because unless you've got a Bloomberg [2:49] terminal." But then what happened? No, [2:51] many people could beat the market [2:53] without a Bloomberg terminal since the [2:54] 1980s. [2:56] And then in the 1990s, [2:58] index funds became very popular where we [3:01] had S&P 500 index funds and people said, [3:04] "Wait, if everyone is buying the index, [3:06] then you you cannot beat the index." And [3:09] then in the 2000s, [3:11] there quant fund quantitative funds [3:14] that again traded based on computers. [3:17] Those were the the early type of high [3:20] frequency trading and people said with [3:21] all these [3:23] computer trading, there's no way you can [3:24] beat them. So, this is not a new [3:26] narrative. This has been said again and [3:28] again every 10 years. Now, the new [3:31] excuse that people give who are not able [3:33] to beat the markets is it's all AI's [3:36] fault. This is absolute rubbish and I'm [3:38] going to break it down step-by-step in [3:41] this video. So, let's go through the [3:42] arguments. What's the first argument? [3:44] With AI, people can analyze any stocks [3:47] so fast so you can't find any more [3:49] undervalued companies among great [3:51] businesses. Absolute rubbish. [3:53] So, for example, if you go to Stock [3:55] Oracle, click on Value Radar, you can [3:58] see in the S&P 500, are there good [4:00] companies that are still undervalued? [4:03] Yes, they are. Of course, last month [4:05] there were more, now they are less [4:08] because the market has gone up quite a [4:09] bit, but there are still stocks that are [4:11] undervalued. And again, these are [4:12] companies that people can analyze with [4:14] AI [4:16] in seconds. In fact, using Stock Oracle, [4:18] which is an AI-powered tool, you can [4:20] analyze it in minutes and seconds as [4:22] well. So, for example, take a look at [4:24] this heat map. What stocks are [4:27] undervalued right now? Straight away, [4:28] Microsoft. One of the most [4:31] analyzed companies in the world is [4:34] undervalued by 11.98%. [4:37] What else? Meta Platforms, that owns [4:39] Facebook, Instagram, WhatsApp, that's [4:42] undervalued by 34%. Eli Lilly, one of [4:45] the top pharmaceutical companies in the [4:47] world, although I have sworn off drug [4:49] companies, that's 12% undervalued. And [4:52] you've got ServiceNow, of the top [4:54] software companies, [4:56] uh undervalued at about 38% undervalued. [4:59] Of course, let's not forget United [5:01] Health, that is currently 10% [5:03] undervalued. In fact, last year when I [5:04] bought it, it was like 50% undervalued. [5:07] And couple of months ago, Google was [5:10] very undervalued. Nvidia was very [5:12] undervalued just a few weeks ago. And uh [5:16] what else? Uh Netflix was very [5:17] undervalued a few weeks ago. So, that [5:20] argument doesn't hold water. Yes, you [5:22] can find great companies that are [5:24] undervalued even in the age of AI when [5:26] people can analyze stocks very, very [5:28] fast. Second argument people make is [5:30] that with AI, it erodes the edge of [5:34] stock pickers, that you can't make money [5:36] picking stocks anymore because now [5:37] everyone can ask AI what stocks to pick, [5:40] and if everyone picks the same stocks, [5:41] who can make money in picking stocks? [5:43] Total And let me show you the [5:45] proof. The proof is in my performance [5:48] and all my students who are following [5:50] the value momentum investing philosophy. [5:53] So, year to date, you can see that my [5:55] portfolio, which I just transferred to a [5:57] corporate account, [5:59] uh is up 25.66% [6:02] as of the 13th of August yesterday. And [6:05] I'm way outperforming the S&P 500, which [6:07] is up 13% year to date. [6:10] See, I can still beat the market. In [6:13] fact, ever since ChatGPT ChatGPT was [6:16] launched on the 30th of November 2022, [6:21] and you if I show you the chart from [6:23] that day, this is when AI was launched, [6:25] I have outperformed the market even [6:28] more, all right? So, uh my portfolio is [6:31] up 137% [6:33] return versus the S&P 500 that is up [6:36] only 90% return. And of course, you can [6:39] see what happened here in February this [6:42] year, I withdrew 99% of all my money [6:47] from my personal account of my wife, and [6:49] I transferred it to a newly-formed [6:51] private limited account in order to [6:55] uh avoid potential estate tax from the [6:58] US government, right? So, it's more of a [7:00] administration issue. But, the whole [7:01] point is to show you that yeah, [7:03] I'm still beating the market. In fact, [7:05] I'm beating the market even more [7:08] in the age of AI. I'm going to explain [7:10] to you in a short while why in fact AI [7:12] has made it even easier for me to beat [7:15] the market and to find opportunities [7:17] regularly. And of course, it's not just [7:19] me. [7:20] Uh you can read my students sharing [7:22] their testimonials almost every day [7:24] they're posting it. By the way, be [7:26] careful. I just saw on social media [7:27] there are some people who are taking my [7:30] students' testimonials and using [7:31] Photoshop and putting different names in [7:33] there. So, just be careful about that. [7:35] So, this is show one of my uh [7:37] subscribers, and he said, "Hi Adam, uh [7:40] just wanted to send you a personal thank [7:42] you. Looking back at the chart really [7:44] reminds me of how far I've come. I [7:47] joined you on 26th August 2024. And [7:50] again, this was after AI [7:53] has begun, you know? And so, technically [7:55] this overall return could have been much [7:57] better. Before joining you, you can [7:59] actually see from the earlier part of [8:01] the graph over here, you can see, I [8:03] suffered pretty huge losses from my own [8:05] mistakes. [8:07] Uh, after learning your methodology, the [8:09] biggest change wasn't simply what to [8:10] buy, but finding a proper system, [8:13] structure, and rules to follow. Instead [8:15] of jumping from one idea to another. And [8:17] as a result, since then, [8:19] uh, he made 445,000 [8:22] in net profit. His account is up 37% [8:27] uh, since that period. And again, [8:29] in the age of AI, he's beating the [8:31] market. Another one of my students who [8:32] just sent in this testimonial, started [8:35] my investment journey February 2026, [8:38] just this year, enrolled in the Whale [8:40] Investor course, learned to invest in [8:42] the top 1% of stocks, and up 42% [8:47] uh, year-to-date. So, do I use AI to [8:49] pick stocks? Do my students use AI to [8:51] pick stocks? The answer is no. In fact, [8:54] I'm here to tell you that if you use AI [8:56] to pick stocks, you ask AI, "Hey, what [8:58] stocks to buy?" I guarantee you you will [9:01] not do well. You will not make money. [9:04] Or, you will not beat the market. Why? [9:07] Because understand this, AI is very good [9:10] for stock research, research, but it is [9:14] terrible at telling you what stocks to [9:17] buy. And here's the reason. [9:19] What is AI? AI is artificial [9:21] intelligence. It is trained on [9:24] historical data. [9:26] And so, AI is trained on what we call [9:28] consensus information. When you ask AI, [9:32] "What stocks do I do I buy?" where does [9:35] AI give you the answer from? [9:37] It will consolidate [9:39] what everyone already believes. It [9:42] consolidates the majority opinion. So, [9:44] if everyone thinks that this stock is [9:46] good and everyone thinks the stock will [9:49] go up and everyone thinks the stock is a [9:51] big and you pick that stock, guess what? [9:54] There's no more edge. Because if [9:57] everyone thinks they've got an edge, no [9:59] one has an edge. So, essentially, when [10:01] you buy stocks picked by AI, [10:04] you're picking what the crowd believes [10:06] is already true. So, you cannot beat the [10:08] market because AI [10:10] is the market. Okay, finally, let's look [10:12] at the argument that high-frequency [10:14] trading has eroded all the profits in [10:17] the market. That is true if you are a [10:20] very short-term trader. If you are like [10:22] a day trader, it's almost impossible to [10:24] make money because you're competing [10:25] against again these supercomputers that [10:28] buy and sell in microseconds. But, if [10:31] you are an investor with a longer time [10:33] horizon, high-frequency trading actually [10:36] creates more opportunities and I'm going [10:38] to explain why. So, first of all, let's [10:40] take a look at the facts. Currently, [10:42] about 60 to 70% [10:45] of transactions in the market of the [10:47] buying and selling are not done by human [10:49] beings. [10:50] They're done by robots, high-frequency [10:53] robots. [10:54] And these robots, they buy and sell with [10:58] no regards to the underlying [11:00] fundamentals of the business, with no [11:02] regards to the intrinsic value. They buy [11:04] and sell purely based [11:06] on price action. [11:08] And what these robots do to the market [11:11] is that it increases the volatility of [11:14] the market. It makes the market more [11:16] volatile. It makes uh sharper drops and [11:20] sharper rebounds. So, is that good for [11:22] investing? Yes. In fact, it helps us to [11:25] make even more money. And let me explain [11:26] why. So, first of all, let me explain to [11:28] you my methodology of how I make money [11:31] from the markets. So, as you guys know, [11:33] I am known as a value investor. But but [11:36] to be specific, I am a quality [11:40] value investor. In other words, to me, [11:41] valuation is not the most important [11:43] thing. [11:44] The most important thing is to only [11:45] invest in the top 1% of the highest [11:49] quality companies. Valuation is [11:52] secondary, okay? So, in other words, [11:55] by investing in the top 1% of companies, [11:58] these are companies with a strong [12:00] economic moat. These are companies that [12:02] have got consistent growth in revenue, [12:05] profits, and cash flow from operations. [12:08] These are companies with high return on [12:09] equity, high return on capital. And [12:12] again, using Stock Oracle, which is [12:15] which is an AI-enabled tool, you can get [12:17] all this information very very quickly. [12:19] So, for example, let's look at [12:21] Microsoft. [12:23] So, if I punch in Microsoft, uh straight [12:25] away you can see [12:27] uh this gives you a very good [12:29] um summary. In terms of how predictable [12:32] are the revenue and profits and cash [12:34] flow? Very predictable. How profitable [12:37] is the company? Very profitable. Growth [12:40] is medium growth, which is good enough [12:42] for a large-cap company. Moat is very [12:44] strong. Financial strength, very strong. [12:47] So, that is what we want to look at. [12:48] Now, to be specific, you can see here [12:51] it's got a wide moat. So, we want to [12:53] focus mainly on wide moat companies that [12:55] have got little or no competition that [12:58] cannot be disrupted easily by [13:00] competition, okay? [13:02] And if you look at this um [13:05] financial history, [13:08] what we want to see again is a company [13:10] where the revenue is growing very very [13:12] consistently. [13:13] And the operating income is also [13:16] increasing very consistently. [13:18] And the [13:20] operating cash flow, [13:22] net operating cash flow is increasing [13:24] very consistently. [13:26] That's what we want to see. That's a [13:27] high quality business. [13:30] And a business with high return on [13:33] invested capital of at least 12 to 15%. [13:37] It's 26% over here. Return on equity at [13:40] least 12 to 15%. That's 34% over here. [13:43] So, that's a high quality company. And [13:46] again, in the market less than 1% [13:50] of companies pass this criteria. [13:54] Okay? So, why do I only focus on these [13:57] kind of companies? Because these are [13:58] companies where [14:01] I have a high level of conviction that [14:03] the intrinsic value, the value of the [14:06] business will rise over time. [14:10] The business will become more and more [14:12] valuable over time as it generates more [14:14] and more cash from operations. But, will [14:16] the share price go up in a straight [14:18] line? [14:19] No, the share price [14:21] will not go up every day, every week, [14:23] every month. Because in the short term, [14:25] the share price is driven not by the [14:28] fundamentals of the business. In the [14:30] short term, the share price is driven by [14:33] emotions, news, sentiment, and [14:36] manipulation [14:37] by high frequency algo trading. And [14:41] therein lies the opportunity. So, what [14:43] happens is that when there's good news, [14:45] when there's manipulation, what happens? [14:48] The stock PRICE CAN GO WHOA! [14:51] THE PRICE can go above the [14:53] intrinsic value, all right? And then [14:55] when there's bad news, when there's [14:56] manipulation, when there's fear, the [14:59] the stock price can go below the [15:00] intrinsic value. And then you got good [15:02] news, bad news, good news, bad news. [15:05] Okay? [15:06] So, when do I buy a great business? I [15:09] buy it when the price drops below the [15:13] intrinsic value. Make sense? So, in [15:15] other words, this is my buy zone. This [15:18] is my buy zone. This is my buy zone. As [15:22] long as the price is below this green [15:24] line, below the intrinsic value, [15:26] that is where I accumulate shares. Now, [15:29] question, [15:30] must I buy right at the bottom here? [15:33] Must I buy at the bottom of a wave down? [15:36] Must I buy right at the bottom here? No. [15:38] I would love to, but no one can predict [15:41] exactly where's the bottom, right? So, [15:43] we don't have to buy the exact bottom to [15:46] make money. [15:47] We just have to buy near the bottom, [15:50] near the bottom. And how do we know it's [15:53] near the bottom? Again, as long as it's [15:54] below the intrinsic value, [15:57] and if you get a margin of safety, for [15:59] example, you get, you know, 20% [16:02] undervalued, I will start to accumulate. [16:04] I'll buy in tranches at support levels. [16:08] So, I won't go all in at once. I'll buy [16:10] maybe a quarter first, and then if it [16:12] drops a bit more, I buy another quarter. [16:14] If it starts to rebound back, I buy [16:15] another quarter. So, I slowly build my [16:18] position, so my average cost is [16:21] somewhere there, right? Not the bottom, [16:23] somewhere there. My average cost is [16:25] somewhere there. You know, as long as I [16:27] buy somewhere near the low and below the [16:30] intrinsic value, over time, as the share [16:33] price, or rather, as the intrinsic value [16:35] goes up, and and the share price trends [16:38] higher, [16:40] that's how I compound my wealth, okay? [16:43] Now, here's the good thing. [16:46] In the old days, [16:48] before AI, before high-frequency [16:50] trading, [16:52] price reacted very slowly. It took a [16:55] long time for the price to drop below [16:58] the intrinsic value. It took a long [17:00] time. [17:01] Long time, okay? And it also took a long [17:05] time for the price to recover above [17:07] intrinsic value. It took years to do it. [17:11] But now, because of high-frequency [17:13] trading, [17:15] it is very, very fast. And let me show [17:17] you the the data over here. Yeah? So, [17:20] the first data I'm going to show you is [17:21] this, that AI and again machine trading [17:25] have spiked volatility during events. [17:29] So, [17:30] what you see on the x-axis, this is what [17:32] we call the volatility index. How [17:34] volatile is the market measured by the [17:36] VIX? Now, you can see in gray, [17:40] this is before AI, before algo trading [17:44] became popular back in 1998, [17:47] the 2000, 1997, what happened? During [17:51] these major crisis events, the VIX went [17:54] up to [17:56] about close to 50, right? So, it was [18:00] volatile, [18:01] but not as volatile as today. So, what [18:04] happened is during the algo-dominated [18:07] era, which is the ones in blue, when [18:09] algo trading became more and more [18:11] popular, machine trading became more and [18:12] more widespread, you can see 2018, 2015, [18:16] 2025, 2025, 2025, 2025, 2024. [18:21] Now, you can see during major events, [18:24] the volatility the volatility spikes a [18:27] lot more than in the past. [18:30] So, how does this apply to this? It [18:33] means the market gets more and more [18:36] volatile, which is good because the more [18:38] volatile the market, the more we get the [18:41] mispricing, [18:42] and the more I'm able to buy at 30, 40% [18:46] undervalued more frequently. Okay? And [18:50] the other table I can show you is this [18:52] table over here. [18:54] These are the number of days [18:57] it takes for the market to drop to a [19:00] bear market. [19:01] So, a bear market is defined as a drop [19:03] of 20% or more, right? [19:06] You can see that historically, how long [19:09] did it take for the market to drop 20%? [19:12] before AI, before high frequency [19:15] trading. [19:16] In the old days, it took 250 days on [19:20] average. It took almost [19:22] almost three quarters of a year for the [19:24] market to drop 20%. It was a very slow [19:27] process. [19:28] But now because of AI, because of algo [19:30] trading, you find [19:32] that in blue, the algo era, it now takes [19:36] shorter, right? Like in the 2022 bear [19:39] market, the market dropped 20% in just [19:43] over 100 days. During the 2025 tariff [19:46] crash, the market dropped 20% in [19:50] just less than 30 days. And in COVID, [19:53] the market dropped 20% [19:55] in less than less than a month. Okay? [19:58] So, is this good or bad? This is good. [20:01] This is fantastic because it means that [20:03] the market [20:05] you know, the market drops [20:07] very fast, rebounds fast, and you get a [20:09] lot of these V-shapes. [20:11] So, if you can control your emotions, [20:14] and you can focus on great companies, [20:16] and you understand intrinsic value, [20:18] that's how you make a lot of money in [20:20] the market. So, like I said, currently, [20:22] you have got great companies like [20:23] Microsoft that are still undervalued. Of [20:26] course, they were even more attractive a [20:28] couple of weeks ago. It's still [20:29] undervalued, but not as attractive [20:31] anymore. If you look at the chart, you [20:32] can see what I'm talking about, right? [20:34] Wave up, wave down, wave up, wave down, [20:36] wave up, wave down, wave up, wave down, [20:38] wave up, wave down. [20:41] This was a great time to add after wave [20:43] down to support, wave up, wave down. [20:45] It's a double bottom pattern, and now [20:47] it's waving back up. And again, that's [20:49] the intrinsic value over there, 570. [20:51] Still below the intrinsic value, but of [20:53] course, the best time to add is after a [20:55] wave down to support. [20:57] And [20:59] um of course, now that it's waving up [21:01] halfway, it's not as great a time to [21:03] add, of course. But if you You at other [21:05] companies like Meta, I think Meta is [21:08] still within [21:10] a pretty attractive level, right? So, [21:13] Meta's intrinsic value is about $900. [21:16] And again, you can calculate the [21:17] intrinsic value manually, [21:20] or you can look at the Oracle value, [21:22] which [21:24] comes up with a valuation. So, let's put [21:26] in Meta Platforms. [21:30] And you can see Meta ranks very highly [21:32] as well in terms of predictability, very [21:35] predictable. Profitability, very [21:37] profitable. Growth, medium growth. Mode, [21:40] very strong wide mode company. That's [21:43] what we want to see. Financial strength, [21:45] very strong. So, if people are concerned [21:47] that all these companies are taking on [21:49] so much debt, so much debt, so much [21:51] debt, but in perspective, yes, there's a [21:53] lot of debt in billions, but if you look [21:55] at their cash flow, you look at their [21:57] assets, cash and equivalents, it's still [22:00] very manageable. Okay? And again, very [22:03] undervalued. Intrinsic value somewhere [22:05] there, about 900 bucks, and it's again [22:07] it's 34% [22:09] undervalued. [22:11] And so, but I already bought a lot of [22:14] Meta, so I'm not adding more. But what [22:15] I'm saying is if I didn't have enough [22:18] Meta, would I add more here? I wouldn't [22:20] mind. [22:21] If you look at Meta, again you can see [22:24] you know, wave up, wave down, wave up, [22:26] wave down, wave up, wave down, wave up, [22:28] wave down, wave up, wave down. So, very [22:31] strong support here, right? And the girl [22:34] hasn't run away yet. The girl is still [22:36] waiting at the support, and eventually [22:39] this is going to to move up a lot, [22:41] right? So, a lot of opportunities in the [22:42] market. Be patient, focus on the great [22:45] companies, and you do very well. And [22:47] yes, AI [22:49] is terrible at picking stocks, [22:52] but it's great for research. And of [22:54] course, with Stock Oracle, which is [22:55] AI-enabled, it ensures that you get [22:58] accurate research. Because the problem [22:59] is, if you just go to ChatGPT or Gemini [23:03] and ask them to research a stock, [23:05] remember about 30% [23:08] of the information is hallucinations. [23:10] That means the AI makes up. Anyway, [23:14] this is a fact. If you ask your AI [23:16] chatbot, Gemini, chat GPT, Grok, "Hey, [23:19] how much of your information is [23:21] hallucination?" It'll tell you, "Yeah, I [23:22] make up. In fact, I asked Claude [23:26] how much of the data it gives me from a [23:28] prompt is hallucination. And it's very [23:32] honest. It It says over here, right? [23:33] Hallucinated data, chatbots state wrong [23:37] earnings per share, stale prices, and [23:40] and invented ratios with total [23:43] confidence. I'm right. [23:45] And [23:46] 39% of the time, [23:49] it tells you the wrong thing. And this [23:52] is your hard-earned money. How the hell [23:54] can you trust your hard-earned money to [23:56] an AI that gives you 39% made-up [24:01] This reminds me of this very funny [24:02] cartoon where this guy asked the AI, "Is [24:05] this mushroom edible?" Yes, with a lot [24:07] of confidence. And then the guy dies and [24:09] the AI GOES, "YOU'RE RIGHT. It's a [24:11] poisonous mushroom. Would you like to [24:14] learn more about poisonous mushrooms?" [24:16] Well, no. I'm dead. [24:19] If you want to catch my latest videos, [24:20] click on the subscribe button right now. [24:23] Click on the bell so you get instant [24:24] notifications once I upload my latest [24:27] video. If you want to check out my [24:29] online course, also go to [24:30] piranaprofits.com. [24:32] We're going to learn how to invest and [24:34] how to trade the financial markets and [24:36] create an income from all around the [24:37] world. [24:39] If you want to join my live wealth [24:40] academy program, go on to [24:42] wealthacademylive.com [24:44] and find out more about how you can [24:45] learn investing and trading live online. [24:48] This is Adam Khoo, and may the markets [24:50] be with you.