Video summary
Broad Market EXPANDS! Time To Rotate OUT of AI Stocks?
Main summary
Key takeaways
Market recap (macro + rotation)
-
Broad market broadened / equal-weight outperformance
- S&P 500 (S&P): down ~4% (as stated)
- NASDAQ-100 (QQQ): down ~1.52%
- Equal-weight S&P (RSP): up, reaching a new all-time high → interpreted as rotation away from concentrated AI/growth winners toward a wider set of sectors.
-
Key sectors participating
- Strong: Financials (XLF), Healthcare, Communication Services, Real Estate, cyclicals, and some consumer staples/defensive
- Weak: Semiconductors / AI infrastructure
-
Semis hit hard
- SMH: down 5.4%
- Narrative driver: Meta-related AI compute headline (detailed below).
Major news driving intraday moves (with explicit implications)
1) Meta “cloud push” / AI compute resell headline
- Meta (META): gained strongly (up almost 9%, one of its best days)
-
Headline theme (Bloomberg; described as unconfirmed by Meta)
- Meta considering selling access to AI models hosted on its infrastructure and/or selling raw compute capacity
- Framed under its MetaMP initiative
- Context: Meta supposedly has excess AI compute capacity and may be leveraging a cloud/computing business
-
Market reaction (per commentary)
- Triggered algorithmic sell-offs across AI infrastructure-related names (semis / AI infrastructure)
- Described as “first-derivative” trading from the headline
-
Specific agreements / counterparties mentioned
- CoreWeave (CORZ): Meta referenced as having a ~$20B deal
- Commentary cited a caution/qualification that Meta may not have the right to resell leased compute capacity from CoreWeave through 2032 (based on analyst checks)
-
Counter-argument presented (bullish/neutral framing)
- The compute constraint remains real; selling some excess compute wouldn’t necessarily negate broader demand
- “Backlog proof points” cited:
- Amazon: $265B backlog
- Google: $400B backlog
- Oracle: $600B backlog
- Microsoft: $500B backlog
- CoreWeave: $100B backlog
- Conclusion: headline likely causes consolidation, not a thesis break on AI demand
2) Amazon designing custom AI chips (on-device compute)
- Amazon (no ticker) designing AI chips for Echo, Fire TV, and future devices
- Mentioned roadmap:
- October: unveiling A3 and A3 Pro, designed for on-device AI models vs cloud
- Implication given: potentially negative for Qualcomm (QCOM) (on-device chip ecosystem)
3) Microsoft job cuts (AI as scapegoat narrative)
- Microsoft (MSFT): another round of job cuts next week
- Expected to affect <2.5% of 220,000 employees
- Expected impact: sales, consulting, and Xbox roles (per subtitles)
- Commentary framing
- AI cited as justification, but underlying driver suggested as over-hiring in 2020–2021
- Not necessarily bearish for AI investing
4) OpenAI inference optimization (compute cost reduction)
- OpenAI engineers found inference optimizations:
- “More than half” the cost of running existing models
- For ChatGPT traffic (without free/paid accounts), at one point only a couple hundred Nvidia GPUs needed
- Caution noted: Jevons paradox
- If inference gets cheaper, usage/token consumption may increase, potentially offsetting cost reductions
5) Apple seeking Chinese memory chips (national security constraint risk)
- Apple seeking to buy Chinese-made memory chips
- Suppliers mentioned:
- CXMT
- YMTC
- Note:
- Both reportedly on the Pentagon warning list of Chinese companies believed to support Beijing militarily
- Uncertainty whether purchases are allowed
- Speaker reiterates: memory remains constrained for the next few years (tied to earlier reports on Micron)
6) OpenAI proposes handing Trump administration a 5% stake
- OpenAI proposes a 5% stake to the Trump administration
- Valuation range cited:
- $850B to $1T (as stated)
- Framed as potential alignment / national security governance angle
7) Robotics / robotics strategy (geopolitical)
- Boston Dynamics CEO urges a national robotic strategy to counter China
- Claim cited:
- China accounts for 54%+ of global robotic deployments
- Broader theme mentioned:
- supply chain / safety / trust
- speaker notes humanoid robotics momentum and training visibility in China
Economic data & rates focus
- JOLTS: 7.6M (above estimate; released yesterday per speaker)
- ADP employment change: 98,000 vs 113,000 expected (softer)
-
Non-farm payrolls (tomorrow): emphasized as the key catalyst
- Recommendation/caution: should land within ~10–15% margin of error (based on prior algorithm behavior)
- If too strong: market fears higher rates for longer → 10Y/20Y yields spike
- If too weak: bad for economy/equities
-
Yields
- 10Y/20Y described as still in a daily downtrend
- “Not much notable movement” recently
Commodities / inflation proxies mentioned
-
Oil
- ~$67.5/barrel, described as weak
- Suggested effect: macro “relief” via lower consumer/transportation/input costs
-
Gold
- Struggling below $4,000
- Needs a daily close above the 12 EMA and above $4,085 to confirm a swing long
- Target path: reclaim $4,500 over time
-
Silver
- Similar setup; targets hinted as mid-to-high $60s if tracking gold
-
Crypto (crypto not favored)
- Bitcoin
- Support/range around $59,000
- Break above $615 (12 MA mentioned) could trigger a swing long, but framed as counter-trend due to higher timeframe bearishness
- Resistance zone: $66 to $73
- Ethereum
- Support: $1,750 down to $1,300
- Resistance: $2,000 to ~$2,500
- Speaker: not a favorite trade
- Bitcoin
Technical levels & risk management framework (explicit “lines in the sand”)
Index / broad market levels
-
S&P 500
- “Line in the sand”: ~725
- Bullish structure: “bull flag,” consolidation above EMAs; RSI cooled
-
QQQ
- Daily line: ~702
- Weekly line: ~687
- Additional risk trigger: crack 695 could be a problem
-
Thesis condition
- If semis weaken but Big Tech holds/resilient, losses may be mitigated via rotation
Semiconductors (SMH + specific names)
-
SMH daily 12 MA
- Not lost; still defended
-
Bear triggers
- Daily: break below ~605 → daily downtrend
- Weekly: break/lose weekly levels ~556 (described as far away)
-
Tone
- Headlines causing volatility, but not “major thesis breaking.”
-
TSMC (TSM)
- Line in sand: ~$421
- If broken: potential weekly consolidation
- Next reference: 12 EMA
- “Striking zone” in lower $400s during July
Selected “big tech” stock levels & recommendations
-
Apple (AAPL)
- Constructive after memory-related hit
- Not adding for long-term at current price, but acknowledges swing opportunity
- Long-term re-add requires regaining higher timeframe levels (mention: above ~302)
-
AMD (AMD)
- “Line in the sand”: ~$500
- Higher weekly lows around $440; 12 MA touch ~$440 area mentioned
- Not buying now; holding existing position (~10% valuation as stated)
- Long-term conviction: potentially $1.2T company (not immediate)
-
NVIDIA (NVDA)
- Support range: $200 to $180
- Plan: sell weekly puts, accumulate shares on support
- Valuation guidance: “fair value” $250–$270
-
Meta (META)
- Resistance band: ~$626 to ~$650
- Support repair: below 600 referenced; also ~605 to ~560 area for daily repair
- Recommendation: continue to like buys under/around 600 (support-based adding)
-
Microsoft (MSFT)
- Adds down at structural support
- Uses LEAP calls and selling puts (risk management via strike placement)
- Valuation cited: around 22x forward earnings and ~17% EPS growth
- Framed as “under 20x next year” (unusually cheap)
-
Alphabet (GOOGL)
- Support: ~$350 down to ~$340
- Not adding near $360–$370; valuation concern described (above 2 PEG)
- Possible lower-300s opportunity if it breaks further
-
Netflix (NFLX)
- Support: ~$70 down to ~65
- Swing reversal trigger: close above 76
- Target resistance: 82–86
-
Tesla (TSLA)
- Near-term resistance: ~$430 down to ~415
- Catalyst constraints noted; expects “rising tide” if Mag 7 improves
-
Palantir (PLTR)
- Key reclaim level: ~$130 (bull level)
- Plan: bid/add ~$130 down to lower $100s
- Valuation commentary: ~1.75 forward PEG for the next three years (as stated)
-
SoFi (SOFI)
- Needs weekly break above ~$180–$181.9
- Speaker holds/adding long-duration; mentions LEAP calls bought around ~$16
-
Amazon (AMZN)
- Targets: upside follow-through to ~$235
- Valuation cited: ~27x forward earnings, PEG ~1.3
- Note: chart-based but also earlier headline context
-
Uber (UBER)
- “Gool log” range: stuck between ~$81–$75
- Support: ~$69–$63
- Strategy: “wheel” (sell puts near lows)
-
SpaceX (no ticker)
- Down significantly; speaker not initiating new positions
- Reference level: ~$150 (post-IPO first price)
Performance stats / sizing and explicit trades (examples)
-
Repeated “portfolio construction / trade plan”:
- Use put selling around “structural support” areas
- Use LEAP calls for long-term exposure (examples: MSFT, SOFI)
- Prefer waiting for confirmation vs chasing when momentum turns
-
Example valuation snapshots mentioned:
- NVDA: ~20x forward earnings, under 1 PEG at ~$195
- MSFT: ~22x forward earnings, ~17% EPS growth; “~under 20x” next year
- NFLX: ~20x forward earnings, ~1 PEG at ~$75
- AAPL: not adding long-term due to “price too high,” but potential swing from lows
- GOOGL: not adding because valuation extended (above 2 PEG)
- PLTR: sub-130 described as ~1.75 forward PEG over the next three years
-
Example ETF/sectors:
- XLF: targeting follow-through; next stop ~$56.50 (former all-time highs referenced)
- Insurance ETF: breakout described; bullish candle/structure
Explicit recommendations / cautions
-
Don’t treat the Meta headline as thesis-break evidence against AI demand
- Sell-off described as overreaction
- Suggested outcome: consolidation, not an AI unwind
-
Watch semi consolidation risk
- Semis still strongest, but speaker expects higher-timeframe consolidation possible in July
- Trade adaptation: broaden exposure to other sectors if semis pause
-
Macro caution tied to jobs report
- NFP should be near in-line (±10–15%) to avoid yield shocks and equity detraction
Tickers / instruments mentioned
Index / ETFs
- S&P 500 (index; no ticker), QQQ, RSP, SMH, XLF, XBI
Stocks / ADRs
- META, NVDA, AMD, AAPL, GOOGL, MSFT, AMZN, QCOM, TSM/TSMC (TSM)
- PLTR, NFLX, TSLA, SOFI, UBER
- CoreWeave (CORZ)
Semiconductor / memory / related (partial list)
- Micron (MU), SanDisk (SNDK), Broadcom (AVGO), Qualcomm (QCOM), ASML (ASML)
- AMAT (AMAT), LRCX, INTC, ARM
- Marvell (MRVL), CIFR (“CIFR” / neo-cloud play mentioned)
Financial / exchanges / insurance (partial list)
- SPGI, Moody’s (MCO), Visa (V), Mastercard (MA)
- CME, ICE, plus NASDAQ exchange and CBOE (exchange tickers not specified)
Other named (partial list)
- Reddit (RDDT), AppLovin (APP), Celsius (CELH), Robinhood (HOOD)
- ServiceNow (NOW), Salesforce (CRM), Adobe (ADBE)
- Axon (AXON), Zeta (ZETA), Elf Beauty (ELF), Nike (NKE)
- SpaceX (no ticker), Rocket Lab (RKLB), Iridium (named; not a ticker)
- Bitcoin (BTC), Ethereum (ETH)
- Caterpillar (CAT), Vertiv (VRT), Bloom Energy (BE)
- Nebius (ticker not specified), plus various other briefly mentioned names
Presenters / sources mentioned
- Presenter/Host: “Will” (repeated reference “Will back here…”)
-
External sources:
- Bloomberg (Meta compute initiative; also Apple memory-chip reporting)
- Financial Times (earlier Apple memory-chip context)
- CNBC (Amazon chip comments; OpenAI stake discussion context; Alex Karp mention)
- Rosenblatt (notes/checks on Meta mentioned)
-
Named individuals:
- Alex Karp (Palantir; CNBC speech referenced)
- Mark Zuckerberg (Meta earnings call reference)
- Boston Dynamics CEO (robotics strategy remarks referenced)
- Eli Lilly (appears in subtitles; ticker not stated)
Disclosures / disclaimers
- Subtitles include a promotional/Discord pitch and a general “full disclaimer” line for the presenter’s options positioning (e.g., holding LEAPs).
- No clearly visible explicit “not financial advice” text was included in the provided subtitles excerpt.