Last Update 04 Aug 26
Fair value Decreased 21%Revenue Grew 28%. Free Cash Flow Fell 91%. Cancel the Limit Order.
Cancel the Order. Not Because the Company Got Worse. Because My Number Was Wrong.
In April I told you to set a limit order at $435 and read a book. In May I raised it to $555, told you the company had gotten materially better, and suggested a different book.
Today I am telling you to cancel the $555 order. The new number is $440.
Sit with that for a moment, because it deserves the discomfort. In four months I have marched you from $435 up to $555 and back down to $440, and the stock itself has done almost nothing except wander from $536 to $608 and back to $590.24, where it closed on August 3. If your order filled on some intraday flush I did not track, what follows matters even more, not less. For everyone else, the cost of my May enthusiasm was zero dollars and one unused limit order.
The May revision was built on a spreadsheet assumption that the last ninety days quietly took out back and shot. This note is about which assumption died, what killed it, and why the answer to "do I buy now" is still the same word it was in April.
Rule one of a thesis update has not changed. You do not let the price tell you what to think about the company. You let the company tell you what to think about the price. What is new this quarter is that the company told me something I did not want to hear about my own model.
*Figure 1. META daily close, May 1 to August 3, 2026, ending at $590.24. The May target buy price of $555 and the revised August target of $440 are overlaid as horizontal lines, with the June high and the July 29 earnings release annotated.*
---
The Best Ad Business on Earth Just Produced $784 Million of Free Cash
Start with the part that went right, because plenty did.
Q2 2026 revenue came in around $60.8 billion, up 28% year over year. Slower than Q1's 33%, and still the fastest growth of any large advertising business in the world. The composition is the healthy kind: ad impressions up 14% and price per ad up 12% at the same time. Volume and price rising together is what pricing power looks like when it is real. The AI ad engine is not a story on a slide anymore. It is in the numbers, two quarters running, with no tax gimmick or currency tailwind to explain it away.
There is even the beginning of a second revenue line. Family of Apps "other" revenue, the closest public proxy for enterprise AI and compute, hit $1.0 billion in the quarter, up 73% year over year. Management says external offers for its compute come in "at a significant premium." Small numbers, real slope. The reported talks with Anthropic about compute capacity remain unconfirmed, so file that under rumor, not thesis.
Now read the cash flow statement.
Free cash flow for the quarter: $784 million. Down 91% from a year ago. Not billion. Million. A company with roughly $228 billion in trailing revenue and the best ad machine ever assembled kept, after capital expenditure, about what a mid-cap regional bank earns in a quarter.
And the funding of the gap has changed character. Buybacks have stopped. Long-term debt went from $58.75 billion at the end of March to $83.7 billion at the end of June. Twenty-five billion dollars of new bonds in ninety days. For context, that debt figure was $28.8 billion at the end of 2024. It has nearly tripled in eighteen months. The net cash position I called a fortress in April, $22.4 billion as of the May note, is now $6.6 billion. The company kept paying its $2 a share dividend the whole time, which is a choice I would describe as cosmetic. The dividend is not the story. The borrowing is.
In April I wrote that "Meta can afford it" is the most dangerous sentence in capital allocation. This was the quarter the affording became visible. Capex guidance now sits at $130 to $145 billion for the year, against $115.8 billion of operating cash flow in all of 2025. Take the revenue guide of roughly $251 billion and the expense guide of $165 to $169 billion and the arithmetic hands you an operating margin near 34% this year, down from 41% last year. The depreciation from $225.7 billion of plant and equipment is still mostly ahead of the income statement, not behind it.
So the quarter was not beat-and-raise, and it was not beat-and-spend-more. It was something starker. It was the best operating quarter in the company's history and the worst free cash flow quarter in a decade, at the same time, on purpose.
*Figure 2. Quarterly free cash flow, Q1 2025 through Q2 2026. Q2 2026 shows $784 million against $12.39 billion in Q1 2026 and roughly $8.7 billion in Q2 2025. Annotations mark the buyback halt and the $25 billion bond raise. The revenue line above the bars keeps rising while the bars nearly vanish.*
---
You Cannot Taper What You Have Already Signed For
Here is the disclosure that broke my May model, and it did not come from a press release. It came from the fine print of the quarterly filing.
Meta is carrying approximately $182.9 billion in future lease obligations and approximately $237.7 billion in non-cancelable purchase commitments. Add those together and you get roughly $420 billion of contractually committed future spending. Against that, the balance sheet holds $90.3 billion in cash and securities.
Read those numbers again, then go back and read what I wrote in May. My revised $740 base-case fair value assumed capex would taper to $80 billion by 2030 as custom silicon absorbed the workload. That taper was the engine of the whole valuation. It was also, I can now see, a hope wearing a spreadsheet costume. You cannot taper what you have already signed for. The commitments schedule says the buildout is not a phase the company is passing through. It is the business model now, locked in with counterparties who do not care what my DCF assumed.
This is the difference between the April question and the August question. In April the question was whether $135 billion of capex would earn a return. In August the question is what the company is worth given that spending at this scale is contractual, the returns arrive late if they arrive, and the funding has moved from cash flow to the bond market.
To be precise about what did and did not fire from my May trigger list: I said a capex revision higher without revenue acceleration would push my target down. The floor of the guide did rise, from $125 to $130 billion, but revenue accelerated hard, so by the letter of that trigger, no alarm. I also said two consecutive quarters of genuine AI-driven revenue acceleration would let the required discount compress. That one arguably fired in the bulls' favor. And yet my fair value went down, not up. The reason is simple and a little humbling. My triggers were written about the income statement. The quarter's real news was on the cash flow statement and in the commitments footnote, places my May triggers never thought to look.
*Figure 3. Horizontal bars comparing contractually committed future spending (approximately $237.7 billion of non-cancelable purchase commitments plus approximately $182.9 billion of future lease obligations, per the Q1 2026 10-Q) against $90.3 billion of cash and marketable securities and $83.7 billion of long-term debt. The committed bars dwarf the liquidity bar.*
---
The Footnote Grew Teeth
In April the youth-safety litigation was a paragraph in the regulatory section. In May it was a trigger on a watch list. In the June quarter it became a $2.4 billion charge against earnings, and the company's own filing language now warns that the scheduled US trials "may ultimately result in a material loss."
Media reports say the first Instagram trial is underway in Tennessee. I flag that as reported rather than verified, because the sourcing is thin. What is not thin is the charge itself. Companies do not accrue $2.4 billion for decoration.
My May trigger said I would move if the courts established a meaningful per-user damages framework. A charge is not a framework. But a $2.4 billion accrual plus a "material loss" warning is the sound of a footnote clearing its throat. My invalidation line now sits at aggregate youth-litigation charges above $15 billion, or a European structural ban on the ads model itself. Neither has happened. Both are now closer than they were in May.
The FTC appeal, for what it is worth, still sits in the appellate pipeline with a decision window around 2027. The trial court already rejected the breakup theory. That piece of the May analysis survives intact.
---
The Market and I Now Agree to the Decimal. That Is Not Comfort. That Is a Warning.
Now the valuation, and the apology that comes with it.
This time the work ran deeper than either previous note. Before building any scenario of my own, I inverted the problem and asked what the current price already assumes. At $590.24, holding my base margins and capex path, the market is paying for a ten-year revenue compound growth rate of 11.4%. My own base case, built independently from guidance, base rates for mega-caps crossing $150 billion of revenue, and the fade evidence for this sector, lands at 11.2%.
The market and I agree almost to the decimal. When you and the market disagree, one of you is wrong and there is money in finding out which. When you and the market agree, there is nothing on the table. The 11.4% the price implies is fair. Not conservative, not heroic. Fair. And fair is exactly what you should never pay, because fair leaves no compensation for being wrong.
The updated scenarios, at a 9.31% cost of capital:
Bear case. Ad growth decelerates, the capex returns disappoint, the legal and regulatory tail bites. Revenue compounds at 6.7%, margins stay near 33%. Fair value: $292 per share.
Base case. Advertising compounds at 11.2% with a fading rate, margins recover from this year's 34% back to 41% by the mid 2030s, capex normalises to 22% of revenue over a decade rather than five years. Fair value: $586 per share.
Bull case. Enterprise AI becomes a true second revenue line by 2028, margins reach 44%, growth compounds at 14.3%. Fair value: $851 per share.
Weight those at 20% bull, 55% base, 25% bear and the expected value is $566, about 4% below the market. The asymmetry has gone negative: 44% of upside to the bull case against 51% of downside to the bear. For the first time in this series, you are being offered less than a dollar of upside for every dollar of downside.
Two disclosures, because you deserve to know how soft the ground under any of these numbers is. First, roughly three quarters of the base-case value sits in the terminal period, which is a polite way of saying the model's conviction lives mostly beyond 2035. Second, the single largest swing in the whole valuation is not my growth forecast. It is the textbook convention used for beta in the discount rate, worth about $45 per share on its own, which is more than ten times the value of my entire business forecast disagreement with the market. Anyone quoting you a fair value on this company to the nearest dollar is selling precision they do not have. I quote mine to remind you not to trust it too much.
Where does that leave the May numbers? The $740 was wrong, and it was wrong for an identifiable reason: it assumed the spending was temporary, and the commitments schedule now says it is structural. Strip out the phantom capex taper, hold the spending at what the company has actually signed for, and the base case lands at $586. The required margin of safety for a business of this calibre, carrying this risk profile, remains 25%.
Updated Target Buy Price: $440 per share.
The April note said $435. It took me four months and one embarrassing detour through $555 to get back within five dollars of where I started. I would call that a coincidence, except it is not. The April number was set when the market still feared the capex. The May number was set when I briefly believed the capex would politely go away. The August number is set on the evidence that it will not. The evidence and the original instinct agree.
*Figure 4. Three vertical bars (Bear, Base, Bull) showing the August fair-value range, with the current price overlaid as a dashed line at $590.24 and the new target buy price band at $440. A faded reference marker at $740 shows the May base case that the commitments disclosure retired.*
---
What Did the Insiders Do While the Company Borrowed $25 Billion
Short section, because the signal is unchanged and I refuse to pad it.
The named officers remain routine sellers under prearranged 10b5-1 plans. No open-market purchases worth reporting. Insider ownership overall stands near 13.4% of the company, though nearly all of that is the founder's block, and the founder is not buying or selling his control, he is exercising it. The people with the clearest view of the AGI program continue to diversify away from it on schedule.
In May I judged management harshly, largely on the selling and the $9.46 trillion moonshot compensation package. On reflection, and with another quarter of evidence, some of that harshness was misplaced. The selling is programmatic and unremarkable. What management has actually done with the core business this year, growing it 28% at this scale while holding the ad auction's pricing power, is elite operation. My quarrel is not with their competence. It is with the board's decision to fund a contractually locked buildout from the bond market while telling shareholders the free cash flow will come back later. It probably will. "Probably" is why the margin of safety is 25% and not 10%.
---
What Would Actually Make Me Move
Written in advance, as always, so I cannot rewrite them when my pulse rises.
Three things would raise the target toward the quote.
First, the FY2027 capex frame. The next earnings call, expected in late October, will carry the first real signal on 2027 spending. If the guide shows capex growing slower than revenue, the normalisation my base case assumes starts arriving on schedule, and the discount can compress.
Second, the enterprise line getting real. Family of Apps other revenue crossing a $6 billion annualised run rate, or disclosed compute and API contracts with named counterparties, would mean the second revenue line is fact rather than proxy. Meta Connect on September 23 is the venue where that story would surface first.
Third, company-defined free cash flow reappearing. Two consecutive quarters back above $10 billion while the guide holds would prove the collapse was a funding choice, not a trap.
Three things would push the target down or kill the thesis outright.
FY2027 capex guided above $170 billion while the enterprise proxy is still below $6 billion annualised. That combination means the spending compounds and the payback does not.
Ad revenue growth below 10% for two consecutive quarters. The entire structure rests on the ad machine funding the gamble. If the machine slows while the commitments stay signed, the bear case stops being a scenario and starts being a forecast.
Aggregate youth-litigation charges above $15 billion, or a European structural prohibition on the ads model. The $2.4 billion accrual is the first tooth. Fifteen billion would be the jaw.
---
What Did Not Change
The competitive position is intact. 3.6 billion people a day, network effects nobody has replicated, a return on invested capital of 25.8% on a conservative basis that charges the company for its leases and its stock compensation. One honest caveat, though. I looked harder this quarter at the evidence on persistence, and even advantages this wide fade toward the mean faster than their owners believe. The AI-native attention rivals are not hypothetical. Durable is not the same as permanent, and I hold this view with slightly less swagger than I did in April.
Reality Labs lost $19.2 billion last year and is guided to do it again. Three notes into this series and that sentence has never once needed editing.
The balance sheet is still strong in absolute terms, $90.3 billion of cash and securities against $83.7 billion of bonds, interest covered more than forty times over. But "fortress" was the April word, and words should mean things. A fortress does not visit the bond market for $25 billion in a single quarter. Call it a well-capitalised construction site.
The recommendation has not changed since April, through a stock that rose seventy dollars and gave most of it back, and a target of mine that rose $120 and gave nearly all of it back. Wait. Hold if you own it. The waiting is still the work.
---
Set the Lower Order. Apologize to Nobody. Read a Book.
Cancel the $555 order. Set the new one at $440. That is a 25% margin of safety against a $586 base case that finally prices the spending the company has actually signed for, rather than the spending I hoped it would settle for.
If the order fills, you own one of the great business machines of the age at a price that pays you for the capex lock-in, the litigation wave, the Reality Labs subsidy, and the possibility that my base case is still too kind. If it never fills, you have lost nothing but the pleasure of owning what everyone else owns at the price everyone else paid.
The walk-away levels: the stop sits near $352 if a position is ever established. Below $292, the bear case is no longer a scenario, it is the realized world, and you should want no part of averaging into it.
One last thing about the detour. Between April and August my target went $435, $555, $440. An investor who followed all three notes did precisely nothing the entire time and lost precisely nothing. That is not an accident of the framework. That is the framework. The discipline did not prevent the mistake in May. It prevented the mistake from costing anything. That distinction is worth more than any fair value estimate in this series.
Set the order. Read a book. I recommend a long one.
---
*DISCLAIMER: This analysis is dated August 4, 2026 and reflects information available as of that date. It constitutes no investment advice. All financial data and company developments must be independently verified before any capital allocation decision.*
Zuckerberg Is Building a Railroad. The Question Is Whether Anyone Needs to Go Where It Leads.
The pattern never changes. A man builds something extraordinary, generates more cash than God, and then decides that what made him rich isn't interesting enough anymore.
Meta Platforms prints $200.97 billion in annual revenue. The Family of Apps (Facebook, Instagram, WhatsApp, Messenger) reaches 3.58 billion people every single day. That's not a business. That's a tax on human attention. The Advantage+ ad-targeting algorithm has achieved what every monopolist dreams of: it knows what you want to buy before you do, and it charges advertisers a premium for that knowledge.
And yet Mark Zuckerberg looked at this magnificent cash-generating machine and said, "Not enough."
He's spending $135 billion next year to build something called the Prometheus supercluster, a cathedral of silicon dedicated to achieving Artificial General Intelligence. He recruited 29-year-old Alexandr Wang, former CEO and co-founder of Scale AI, following a $14.3 billion investment in that company, to serve as Chief AI Officer and run a division called Meta Superintelligence Labs. The projects are called "Mango" and "Avocado," which tells you something about the culture.
Now, I don't pretend to understand the technical minutiae of large language models. But I understand incentives, and I understand the physics of capital allocation. When a man who controls approximately 61% of the voting rights decides to redirect the entire strategic direction of a $200 billion revenue enterprise toward an unproven technology frontier, you'd better understand the second- and third-order consequences before you hand him your money.
The Moat Is Five Billion Feet Wide. The Castle Behind It Is on Fire.
The advertising business deserves its reputation.
Three and a half billion daily active users generate a data exhaust so rich that Meta's algorithms can predict consumer behaviour with almost eerie precision. The network effects are self-reinforcing: every new user makes the platform more valuable to advertisers, which funds better products, which attracts more users. It's a flywheel that would make any industrialist weep with envy.
WhatsApp paid messaging crossed a $2 billion annual run rate in Q4 2025. That's significant. It proves Meta can extract enterprise software revenue from a consumer platform, a trick that has eluded nearly every social media company in history.
Free cash flow: $43.59 billion. Cash on the balance sheet: $81.59 billion. Operating margin: 41%. Return on invested capital, by my calculation: north of 20%, even after the distortions of the One Big Beautiful Bill Act, and closer to the mid-30s when normalised for that one-time statutory charge.
Show me these numbers in isolation, and I'd tell you this is one of maybe twenty businesses on Earth worth holding forever. Buy it, go fishing.
But you can't look at these numbers in isolation, because the man running the place has decided to set $135 billion on fire in 2026 to chase a technology that may or may not work. That's the intellectual problem.
Show Me the Incentive, and I'll Show You $19 Billion in Losses
Reality Labs lost $19.19 billion in 2025. On $2.21 billion in revenue.
Nineteen billion dollars. Roughly the GDP of Iceland, spent annually with no prospect of near-term return. That's not an investment. That's a subsidy programme funded by the most profitable advertising business on Earth. If Reality Labs were a standalone company, it would rank among the most spectacular money-losing enterprises since the South Sea Company.
The standard defence is that Meta can afford it. And technically, this is true. The balance sheet holds $81 billion in liquidity against $58.74 billion in long-term debt. The company is net-cash positive. It can absorb these losses without threatening solvency.
But "can afford it" is the most dangerous phrase in capital allocation. Wealthy enterprises routinely convince themselves that because they *can* waste money, they *should*. The question is never whether you can survive the loss. The question is whether the capital deployed in Reality Labs would generate superior returns if allocated elsewhere: aggressive share repurchases at current valuations, a special dividend, or simply sitting in Treasury bills earning 5%.
The opportunity cost of $19 billion in annual losses is staggering. And now Zuckerberg wants to layer another $135 billion in infrastructure spending on top of it.
When a 29-Year-Old Runs Your Most Important Division
The formation of Meta Superintelligence Labs under Alexandr Wang is the single most consequential organisational decision in Meta's history. More important than the Instagram acquisition. More important than the pivot to mobile. More important than the metaverse rebrand.
Wang's mandate is breathtaking: build a multimodal image and video model (Project Mango) and a reasoning-focused large language model (Project Avocado) that can compete with, and eventually surpass, OpenAI's frontier capabilities.
In March 2026, Meta quietly executed what I'd call the "Saba Maneuver." A veteran executive named Maher Saba was given control of a parallel "Applied AI Engineering" unit reporting directly to CTO Andrew Bosworth. This effectively split the AI operation in two: Wang gets pure research, Saba gets the budget for products that actually make money.
This tells you everything you need to know about the internal tension. Zuckerberg wants superintelligence. Wall Street wants revenue. The Saba Maneuver is the compromise, and compromises in technology development rarely produce the best possible outcome in either direction.
When a company bifurcates its most important initiative between a visionary and a pragmatist, what you typically get is neither vision nor pragmatism. You get a committee.
The Government Always Shows Up Eventually. The Only Question Is How Hungry It Is When It Arrives.
The regulatory environment has shifted from nuisance to existential threat. Not immediately. These things move slowly. But the trajectory is unmistakable.
Australia banned social media for children under 16. The fines run to A$49.5 million (approximately US$34 million) per violation. If this framework cascades through Western democracies, and the political incentives strongly favour exactly that, Meta's user acquisition pipeline gets structurally capped. You cannot compound Daily Active People when governments are systematically removing the youngest cohort from your platform.
The Australian Competition and Consumer Commission is actively litigating Meta over misleading cryptocurrency advertisements. This isn't a slap on the wrist. It's an assertion that Meta bears algorithmic liability for the content its systems promote. If that legal theory prevails, the implications for the global advertising model are severe.
Meanwhile, Meta's refusal to renew commercial deals with Australian news publishers under the News Media Bargaining Code has triggered new legislation threatening annual penalties equivalent to 2.25% of Meta's Australian-derived revenue. The standoff is escalating, not resolving.
The European Union's AI Act imposes compliance costs and operational restrictions on generative model deployment. If Meta is forced to silo user data by geography or age cohort, the algorithmic efficiency that commands premium ad pricing doesn't merely decline. It disintegrates.
In the United States, a bipartisan coalition of 42 attorneys general filed suit in October 2023, alleging Meta deliberately designed its platforms to addict children. That litigation has since metastasized into a multidistrict proceeding encompassing over 1,700 individual claims. In March 2026, two separate juries found Meta liable for child safety failures: a $375 million penalty in New Mexico and a $6 million negligence verdict in California. These were the first successful jury trials holding a social media company accountable for harm to minors. Courts are also increasingly ruling that Section 230 does not shield Meta from claims about addictive platform design, effectively narrowing the company's legal protections without waiting for Congress to act.
The regulatory contagion is spreading faster than most analysts appreciate. Denmark reached a cross-party agreement in November 2025 to ban social media for children under 15, with implementing legislation being finalized. Brazil's "ECA Digital" took effect on March 17, 2026, requiring guardian-linked accounts for children under 16 and imposing fines of up to 10% of a company's Brazilian revenue for violations. The UK government launched a public consultation in March 2026 on banning social media for children, with the House of Lords twice voting in favour of an under-16 prohibition. Indonesia began enforcing tiered age restrictions on March 28, 2026, deactivating accounts on Instagram, Facebook, and Threads for children under 16. Malaysia's Online Safety Act, effective January 2026, bans social media for under-16s with national ID verification requirements being finalized. What started as one country's experiment is rapidly becoming a global regulatory template.
Nothing is certain except death and taxes. And regulation. Every monopoly eventually attracts the attention of politicians who need someone to blame for their constituents' unhappiness. Meta has 3.58 billion daily users. That's a very large target.
The Tax Bill That Broke Every Quantitative Screener
The "One Big Beautiful Bill Act" (and yes, that is the actual name of actual legislation, signed into law on July 4, 2025) forced a valuation allowance charge that drove Meta's effective tax rate to 30% for the 2025 fiscal year. The normalised rate, absent this one-time statutory distortion, would have been approximately 13%. Meta's own guidance projects a 13-16% effective tax rate for 2026.
This matters enormously for anyone relying on algorithmic screeners or trailing earnings multiples. The Piotroski F-Score appears optically mediocre, with different providers scoring it between 4 and 6 out of 9, because the artificial tax charge depressed reported net income and distorted several of the score's binary criteria. Any quantitative model that fails to normalise for this legislative anomaly will systematically undervalue the enterprise.
The history of investing is littered with fortunes made by people who actually read the financial statements, and fortunes lost by those who trusted their Bloomberg terminal to do the thinking.
A Wonderful Business at a Fair Price Is Still Just a Fair Price
The numbers tell a straightforward story.
Using a 10-year discounted cash flow model with an 8.5% weighted average cost of capital and a 2.5% terminal growth rate, three scenarios emerge:
Bear case. Regulatory fragmentation cripples ad targeting, the compute buildout becomes a stranded asset, revenue growth stalls at 4%, margins compress to 32%. Fair value: $360 per share.
Base case. Advertising sustains a 10% compound annual growth rate through 2030, fading to 5%. Operating margins hold at 40% as Llama edge deployments offset depreciation. The $125 billion median CapEx cycle yields moderate returns. Tax rate normalises to 15%. Fair value: $580 per share.
Bull case. Mango and Avocado achieve commercial dominance, WhatsApp B2B messaging scales exponentially, margins expand to 45% as AI automates internal operations. Fair value: $820 per share.
The stock trades at approximately $536 as of late March 2026. That's a 7.5% discount to the base case.
A 7.5% discount to intrinsic value is not a margin of safety. It's a rounding error. For a business of this calibre undertaking a $135 billion infrastructure gamble while facing hostile regulators across multiple continents, the requirement is 25%.
That puts the target entry at $435 per share.
Patience Is the One Edge That Never Gets Arbitraged Away
The recommendation is simple. Do nothing.
Set a limit order at $435. Wait.
If the market never gives you that price, you haven't lost anything. You've preserved capital for better opportunities. If it does (perhaps because a quarter of disappointing CapEx returns spooks the momentum traders, or because Australian regulators land a particularly painful blow) you acquire a position in one of the world's finest businesses at a price that provides genuine downside protection.
The thesis dies under two conditions: if Daily Active People decline for three consecutive quarters, signalling permanent behavioural shifts or insurmountable regulatory barriers; or if antitrust action fundamentally severs the data-sharing infrastructure between Facebook, Instagram, and WhatsApp. Below $300, walk away entirely.
The limit order exists to enforce discipline, not to manage execution risk.
Everybody Agrees This Is a Winner. That Is Exactly the Problem.
Meta Platforms is executing one of the most audacious corporate pivots in modern history. The financial fortress ($81 billion in liquidity, $43.6 billion in free cash flow, a 41% operating margin) means the enterprise can absorb enormous punishment without threatening its survival. That's rare and valuable.
But the market is pricing in seamless execution of the AGI roadmap while discounting the tail risks. The $135 billion CapEx cycle. The bifurcated AI organisation. The regulatory hostility cascading from Canberra to Brussels. The $19 billion annual bonfire at Reality Labs.
The time to buy Meta is not when everyone agrees it's an AI winner. The time to buy Meta is when a bad quarter or a regulatory shock convinces the market that the entire AGI thesis is dead, and the stock drops to a price where you're being paid to take that risk.
Set the limit order. Go read a book. The best investment decisions tend to involve doing absolutely nothing for extended periods of time.
Have other thoughts on Meta Platforms?
Create your own narrative on this stock, and estimate its Fair Value using our Valuator tool.
Create NarrativeHow well do narratives help inform your perspective?
Comments
3 commentsDisclaimer
tripledub is an employee of Simply Wall St, but has written this narrative in their capacity as an individual investor. tripledub holds no position in NasdaqGS:META. Simply Wall St has no position in any companies mentioned. Simply Wall St may provide the securities issuer or related entities with website advertising services for a fee, on an arm's length basis. These relationships have no impact on the way we conduct our business, the content we host, or how our content is served to users. This narrative is general in nature and explores scenarios and estimates created by the author. The narrative does not reflect the opinions of Simply Wall St, and the views expressed are the opinion of the author alone, acting on their own behalf. These scenarios are not indicative of the company's future performance and are exploratory in the ideas they cover. The fair value estimate's are estimations only, and does not constitute a recommendation to buy or sell any stock, and they do not take account of your objectives, or your financial situation. Note that the author's analysis may not factor in the latest price-sensitive company announcements or qualitative material.
