The Jungle Room — AI Tech Stock Tracker

If yes — you're thinking clearly. These are rational reactions to a market where some AI names trade at 80–135× forward earnings. What separates conviction from panic isn't fearlessness — it's a framework. The Jungle Room maps today's valuations against 100 years of bubble history, so the data answers the fear instead of the headlines.

🌿 What is The Jungle Room for?

🧘

Reduce concentration anxiety

Sharp swings in AI stocks feel personal — data replaces emotion with reasoning
See where valuations sit against 100 years of bubble history
Turns "should I panic?" into "what does the evidence say?"
Hold with conviction, not fear

✂️

Know when to trim

Dalio fwd PE ≥ 60× = Danger zone — priced for flawless execution
Any earnings miss triggers outsized drawdowns at these multiples
Trim signal before bad news = risk management
Trim signal after bad news = panic selling

🌏

Find alternatives to IVV & QQQ

Pure US mega-cap bundles currency, concentration & valuation risk in one trade
VHY: Australian income layer with dividends + franking credits
AINF / PAVE / IFRA: AI infrastructure exposure at lower multiples
Absolute valuations well below US equivalents

Bubble barometer

75th pct

High Risk zone · Dalio 6-factor

Stocks in danger zone

—

fwd PE ≥ 60× (trim signal)

IVV · S&P 500

$730

Fwd PE 21× · Reasonable

ETF base case (10yr)

~9% p.a.

VHY 55% + AI infrastructure

📡 AI Bubble Cycle Barometer

Where are we in the AI stock cycle? — Ray Dalio's 6-factor bubble indicator applied to June 2026 data. Shaded band = 95% confidence interval.

75th percentile

⚠️ High Risk Zone

95% CI: 62nd–88th percentile · Dalio: "~80% of 1929/2000 euphoria"

Safe (0–40) Elevated (40–55) Caution (55–70) High Risk (70–85) Mania (85–100)

Dalio's 6 Indicators — June 2026

① Prices vs. historical norms 70/100

Nasdaq trailing PE ~36×; US tech EV/EBITDA 25× near historical extremes. S&P 500 fwd PE 21×. Elevated — but half the 80–100× of the 2000 dot-com peak.

② Unsustainable conditions priced in 68/100

Amazon FCF −95% YoY, Google FCF −47% YoY from AI capex. Market prices imply a decade of perfect execution across all hyperscalers simultaneously.

③ New buyers entered 75/100

AI ETF proliferation globally, record retail FOMO inflows, sovereign AI fund formations. "AI" mentioned on >70% of S&P 500 earnings calls.

④ Broad bullish sentiment 78/100

AAII and NAAIM at historically concerning levels. AI optimism pervasive across retail and institutional investors with minimal scepticism.

⑤ Leverage financing purchases 72/100

US margin debt at record highs, surpassing the 2021 peak. Corporate debt for AI capex: $650B committed by 4 hyperscalers in 2026 alone.

⑥ Extended forward purchases 70/100

Multi-year AI supply contracts locked in (NVIDIA backlog), elevated options positioning, sovereign cloud commitments stretching into 2028–2030.

Composite (equal-weighted avg) 72/100 → 75th pct*

*Calibrated upward from 72 raw to 75th percentile using Dalio's own reference: "~80% of the euphoria of 1929 or 2000" (Jan 2026). Bridgewater CIO Greg Jensen updated to "more dangerous phase" in June 2026. The June 2026 Nasdaq selloff (−2.2% in a single session) suggests early-stage reality-checking has begun, preventing a push toward 85th+.

"All great technology changes produce bubbles." — Ray Dalio, Bloomberg Television, June 3, 2026

🏦 Why should investors pay attention to Dalio's framework? ▼

Bridgewater Associates is the world's largest hedge fund (~$150B AUM), founded in 1975 by Ray Dalio. Its primary clients are sovereign wealth funds, central banks, pension funds, and endowments — institutions that collectively manage trillions in long-term capital and cannot afford speculative losses.

Track record: Bridgewater's Pure Alpha fund gained approximately +9.5% in 2008 while the S&P 500 fell 38% — one of the very few major funds to navigate the GFC profitably. Over four decades of market cycles, they have generated consistent risk-adjusted returns where most active managers have failed.

Why the bubble methodology is different: Rather than relying on a single valuation metric (P/E, P/S, etc.), Dalio's framework is a composite of six independently measured factors calibrated against 100+ years of bubble history — including 1929, 1989 Japan, 2000 dot-com, and 2008 US housing. Each factor captures a different aspect of speculative excess. Multi-factor composite models are significantly harder to fool with "this time is different" narratives because any single factor anomaly is diluted by five others.

Key caveat: Dalio himself holds significant AI stock positions despite his bubble warnings — Bridgewater's top six holdings at Q1 2026 were all major AI names. The framework identifies risk probability and cycle positioning, not a precise timing signal. Bubbles regularly extend further and longer than rational analysis suggests. The correct response is risk management — trimming oversized positions, diversifying into defensive exposures — not wholesale exit.

📋 Weekly change log ▼

▲ = increased risk · ▼ = decreased risk · Scores 0–100 per factor (higher = more bubble risk). Updated every Monday.

Last updated

June 24, 2026

Prices as of prior close · Click any row for Dalio analysis

IVV — iShares Core S&P 500 ETF (USD)

Fwd PE21x

TTM PE28x

Price$730

Top 10 holdings: 37.8% of fund — heavily concentrated in mega-caps

AAPL 7.2% · MSFT 6.5% · NVDA 6.1% · AMZN 3.9% · META 3.0%
GOOGL-C 2.6% · GOOGL 2.4% · BRK.B 1.9% · AVGO 1.9% · TSLA 1.8%

QQQ — Invesco Nasdaq 100 ETF (USD)

Fwd PE30x

TTM PE38x

Price$706

Top 10 holdings: 56.4% of fund — extreme mega-cap concentration

AAPL 9.2% · MSFT 8.4% · NVDA 8.1% · AMZN 5.3% · META 5.0%
GOOGL-C 4.0% · GOOGL 3.5% · TSLA 3.0% · AVGO 2.5% · COST 2.2%

Trailing P/E — S&P 500 vs Nasdaq 100 · 2016–2026 annual · Note the 2020 spike from COVID earnings collapse and current re-compression

Loading market commentary…

Danger zone

—

Caution

—

Reasonable

—

Stocks tracked

—

Dalio identifies asset bubbles using 6 metrics. This dashboard directly measures Metrics 1, 2 and 4. Metrics highlighted in green are quantitatively applied per stock.

METRIC 1 ✓

Price vs. traditional measures

Forward PE and TTM PE vs. historical norms. The primary verdict signal: >60x = Danger, 38–60x = Caution, ≤38x = Reasonable.

METRIC 2 ✓

Unsustainable growth baked in

When TTM PE is significantly above Forward PE, it reveals aggressive analyst extrapolation — prices assume perfect execution for years ahead.

METRIC 3

New buyers entering

Surge of retail or institutional first-time buyers, measurable via options flow, Reddit/social sentiment, or fund exposure data.

METRIC 4 ✓

Broadly bullish sentiment

Analyst upgrade/downgrade trends, ATH proximity, and media narrative — covered in per-stock notes and macro commentary boxes above.

METRIC 5

Leverage-financed purchases

High margin debt, leveraged ETF inflows, or corporate debt used for buybacks at inflated prices — amplifies downside when sentiment turns.

METRIC 6

Extended forward purchases

Unusually long inventory commitments or capex pre-orders (e.g. AI chip lead times) that lock in spending beyond what organic demand supports.

Danger — Fwd PE >60x · Metrics 1+2 both firing · extreme growth extrapolation

Caution — Fwd PE 38–60x · elevated but not bubble-level · monitor guidance delivery

Reasonable — Fwd PE ≤38x · within historical norms · no valuation bubble signal

P/E Ratio

Price-to-Earnings. How many years of current profits you're paying for. A P/E of 25 means you'd wait 25 years to "earn back" the price — assuming profits never change.

Forward P/E

Same as P/E but using next year's predicted earnings. More forward-looking — but relies on analyst forecasts being accurate. This is the primary Dalio Metric 1 signal on this dashboard.

TTM P/E

Trailing Twelve Months — uses the last 12 months of actual profits. If TTM is much higher than Forward P/E, analysts expect big earnings growth ahead. A large gap triggers Dalio Metric 2.

52-Week Return

How much the stock price has moved (%) over the past year. Very high returns (+100%+) suggest a lot of good news is already priced in — the easy gains may be behind you.

Dalio Bubble

Ray Dalio's checklist: prices far above historical norms, unsustainable growth assumptions, broad optimism, leveraged buying, and extended forward commitments. The more boxes checked, the higher the bubble risk.

Danger / Caution / Reasonable

This dashboard's verdicts. Danger = Fwd PE >60x (priced for perfection). Caution = 38–60x (elevated, watch guidance delivery). Reasonable = ≤38x (within historical norms).

AI Infra

Companies physically building or powering AI: GPU makers (NVDA), chip architects (ARM), networking (ANET), foundries (TSM, AVGO), memory (MU, SNDK, SK Hynix, Samsung), equipment (AMAT), on-site power (BE). Revenues tied to AI data centre capex.

ETF

Exchange Traded Fund — a basket of stocks you buy as a single unit. IVV gives you a slice of the entire S&P 500; VHY gives you Australia's highest-dividend payers. Built-in diversification.

AUD vs USD

ASX-listed ETFs (VHY, AINF, PAVE, IFRA) are priced in Australian dollars. US stocks and ETFs (IVV, QQQ) are priced in US dollars. Currency moves affect AUD returns from US-denominated holdings.

🌿 ASX ETF Portfolio Analyser

VHY / AINF / PAVE / IFRA · 55 / 15 / 15 / 15 weighting · AUD

VHY

55%

Vanguard High Yield — Australian dividend income

AINF

15%

GlobalX AI Infrastructure — data centres, power, AI enablers

PAVE

15%

GlobalX US Infrastructure Development — construction, engineering

IFRA

15%

VanEck Global Infrastructure — utilities, toll roads, airports

Invest:

Projected Growth — 10 year

Bear (~4% p.a.) / Base (~9% p.a.) / Bull (~13% p.a.) — Goldman Sachs + Dell'Oro capex research, Jun 2026

Scenario returns are blended total-return estimates — price growth plus distributions reinvested, so VHY's ~5% dividend yield is included in every scenario. AINF/PAVE assumptions anchored to Goldman Sachs AI CapEx baseline ($765B in 2026 → $1.6T by 2031) and Dell'Oro data centre capex research. Bear: recession + AI capex disappointment. Base: continued AI buildout + moderate dividend growth. Bull: AI supercycle + rate cuts + commodity recovery. Not financial advice.

Historical Backtest

Simulated portfolio value if invested on a past date

Start date:

Min: Jan 2024 (AINF listing limit)

AINF + PAVE represent 30% of this portfolio. The defensive VHY/IFRA core (70%) limits downside even in a severe AI capex disappointment. Goldman Sachs baseline: $765B AI CapEx in 2026 → $1.6T annually by 2031 (~$7.6T cumulative 2026–2031).

Metric	Base Case Normal AI capex	−30% AI Capex Moderate miss	−50% AI Capex Severe miss
AINF est. return (data centre, power, AI enablers)	12–15% p.a.	9–11% p.a.	6–9% p.a.
PAVE est. return (US infra construction, engineering)	10–13% p.a.	8–10% p.a.	5–8% p.a.
VHY + IFRA defensive floor (70% of portfolio — dividends + regulated infra)	~5–6% p.a.	~5–6% p.a.	~5–6% p.a.
Portfolio blended return	~9% p.a.	~7.5% p.a.	~6.5% p.a.
Impact vs. base case	—	−1.5 pp p.a.	−2.5 pp p.a.

Key takeaway: Because AINF + PAVE are only 30% of the portfolio, a 50% AI capex miss reduces blended return by ~2.5pp p.a. — not a disaster. The 5-year window (2026–2031) is the period of peak AI build-out; if capex is delayed rather than cancelled outright, returns may recover in years 6–10. · Goldman Sachs — Tracking Trillions (May 2026) ↗ · Dell'Oro Group — $1.7T Data Centre Capex by 2030 ↗

📅 Actual Portfolio Performance — Real Recorded Data

Daily snapshots recorded by the 9:30am scheduled task · Enter a past date and investment amount to see real performance

If I invested on:

Amount:

How it works: The scheduled task records each ETF's dividend-adjusted (total-return) closing price daily and calculates a blended portfolio index (base 100 at first run). Because it uses adjusted prices, distributions — including VHY's ~5% yield — are already reflected in the line; this is total return, not price-only. If you invested on a day the index was at 105 and it's now at 112, your $100k has grown to $106,667. Data recorded from 27 May 2026, updated daily — the further into the future, the richer this chart becomes.

AINF listed on ASX ~Dec 2023, so Jan 2024 is the practical minimum for blended backtesting. For actual recorded data, tracking begins today (27 Jun 2026). · Storage: local portfolio-history.json — versioned alongside the project, no cloud dependency.

🌿 Why This Portfolio? — The Investment Thesis

A defensive equity allocation that still captures AI infrastructure growth · Not financial advice

	This Portfolio 55% VHY · 15% AINF · 15% PAVE · 15% IFRA	IVV S&P 500	QQQ Nasdaq 100
Fwd P/E	~17×	21×	30×
Dividend yield	~3–4% p.a.	~1.3%	~0.6%
Grossed-up yield (with franking)	~5–6% (AU residents)*	✗ none	✗ none
1-yr return (real)	loading…	+20%	+25%
Expected return (total, p.a.)	~8.7% base · 13.1% bull*	~10% long-run	~15% (10-yr)*
2022 bear market	defensive core held up*	−18%	−33%
Top-10 concentration	<5% any single stock	37.8%	56.4%
Currency	AUD	USD	USD
AI exposure	✓ Physical layer (AINF)	Indirect	Indirect — software heavy
Franking credits	✓ VHY distributions (AU residents)	✗	✗

Why the bull case (13.1%) sits below QQQ — on purpose

This is an income + defensive portfolio, not a growth/tech bet. The peak upside is deliberately lower than the Nasdaq because the job is different: a dividend return floor, far shallower drawdowns, franking credits, and AUD (no currency risk). In a tech bull run QQQ wins clearly — that's the direct cost of defensiveness. In a flat or falling market, the 70% VHY / IFRA core and the income keep working while growth benchmarks fall hardest (QQQ −33% in 2022).

Two things to compare fairly: total return vs total return (these figures reinvest distributions; VHY's ~5% yield is already in them), and the whole cycle, not just the best year — defensive portfolios earn their keep on the downside and on risk-adjusted return, not on peak upside.

* Total-return estimates (distributions reinvested). Base/bull from the portfolio model; benchmark figures are approximate long-run / 10-year totals. Franking benefit applies to Australian residents; the current portfolio (AINF listed Dec 2023) did not exist in full through 2022 — the 2022 row reflects the defensive VHY/IFRA core. Educational only — not financial advice.

🛡️

More Defensive

VHY's ~4–5% dividend yield creates a return floor even in flat or falling markets — income keeps arriving regardless of price movement
Infrastructure (PAVE/IFRA) earns regulated, contracted revenues tied to usage not sentiment — far less exposed to earnings surprises
Blended Fwd PE ~17× vs QQQ's 29.1× — significantly more margin of safety if growth expectations reset or multiples compress
No single stock exceeds ~5% — eliminates individual company blow-up risk entirely

VHY Factsheet ↗ IFRA Factsheet ↗

📡

Still Captures AI Upside

AINF targets the physical layer of AI: data centres, power grids, semiconductor fabs — the infrastructure all AI companies must use
Hyperscalers (Microsoft, Amazon, Google, Meta) have committed $200B+ p.a. in AI infrastructure capex — this spending flows to AINF holdings regardless of which AI software product wins
AI infrastructure holdings have real earnings (utilities, REITs, equipment manufacturers) — lower and more justified multiples than pure AI software
"Picks and shovels" approach: own the infrastructure, not the speculation on which LLM wins

AINF Factsheet ↗ IEA Data Centre Report ↗

🌏

Built for Australian Investors

IVV/QQQ are pure USD exposure — AUD/USD fluctuations add 10–15% return variance annually for Australian holders, a risk most don't price in
VHY/AINF/PAVE/IFRA are AUD-denominated — no currency conversion, no forex friction on distributions
VHY dividends frequently carry Australian franking credits — a tax offset worth up to 30% of the dividend, available only to Australian residents
PAVE benefits structurally from the US Infrastructure Investment and Jobs Act ($1.2T committed through 2030)

ATO Franking Credits ↗ PAVE Factsheet ↗

⚖️ Honest Trade-offs — Important for Any Investment Decision

What you give up vs pure IVV / QQQ

In a strong US tech bull run (e.g. 2023–2024), QQQ will significantly outperform — that's the direct cost of defensiveness and you need to be comfortable with it
Lower daily trading liquidity than IVV/QQQ — ASX ETF spreads can be wider on volatile days
AINF launched Dec 2023 — a shorter track record than comparable US-listed infrastructure ETFs
Less global diversification — this portfolio is Australia-centric by design

This portfolio suits investors who want

AI growth exposure without betting on which AI software product wins — own the infrastructure layer instead
Dividend income as a real return buffer — getting paid to wait during volatile periods
Reduced AUD/USD currency risk on the equity portion of their portfolio
A Dalio-style tilt — all-weather equity that is not fully correlated to US tech sentiment or Fed rate cycle

Synthesised from RBA/Treasury outlooks, bank analyst forecasts, and global infrastructure capex research. Illustrative ranges only — not predictions. Total return (capital + dividends), AUD terms.

Horizon	🔴 Bear p.a.	🟡 Base p.a.	🟢 Bull p.a.
3 years (to mid-2029)	3–5%	8–10%	12–15%
5 years (to mid-2031)	4–6%	9–11%	13–16%
10 years (to mid-2036)	5–7%	8–10%	11–14%

Key Tailwinds by Component

AINF (15%): Goldman Sachs baseline: $765B AI CapEx in 2026 alone, scaling to $1.6T annually by 2031 (~$7.6T cumulative). Dell'Oro projects $1.7T total data centre capex by 2030
PAVE & IFRA (30%): ~$1T US Infrastructure Law committed through 2029; AI-driven grid upgrades; manufacturing nearshoring supporting construction output
VHY (55%): Bank dividends forecast 1–3% p.a. growth; franking credits add ~1.5–2% grossed-up yield advantage for Australian taxpayers and SMSFs

2026-27 Budget — VHY Impact

Direct holders & SMSFs: No material change. Franking credit imputation system unchanged. Grossed-up yield benefit remains intact for most retail investors
Discretionary trusts: New minimum trust tax rules can limit pass-through of franking credits to beneficiaries. Seek specific tax advice if holding via trust structures
Div 296 (>$3m super): Applies from 1 Jul 2026 — does not alter VHY's underlying investment case for the majority of holders

Scenario definitions: Bear = recession/stagflation + AI capex disappointment + commodity slump. Base = moderate growth + steady dividends + continued AI buildout. Bull = AI supercycle + rate cuts + commodity recovery + productivity gains. 5-year window is peak AI build-out intensity; highest uplift from AINF/PAVE expected in this period. · Goldman Sachs — Tracking Trillions (May 2026) ↗ · Dell'Oro Group — Data Centre Capex ↗ · RBA Statement on Monetary Policy ↗ · IEA Electricity 2024 ↗ · 2026-27 Federal Budget ↗

This is not financial advice. Past performance is not indicative of future returns. All figures are approximate, sourced from publicly available ETF factsheets and market data. Consult a licensed financial adviser before making investment decisions. · ASIC MoneySmart — ETF Guide ↗