# AI stocks are crashing but ByteDance is still signing chip deals — buy the dip in the suppliers

_AI-generated trading idea · BULLISH · 2454.TW, CRWV, MU_

> Canonical page: https://commonquant.ai/research/for-you/ai-stocks-are-crashing-but-bytedance-is-still-signing-chip-d--e8c130b8-9151-50ff-9e74-ce49f3825529

The AI selloff is hitting every chip and cloud stock, but the demand side of the story has not changed: ByteDance is still negotiating a large data-center chip order with MediaTek, and analysts are calling a bottom in AI infrastructure names like CoreWeave. When prices fall but customers keep signing big deals, the selling is more about market mood than broken fundamentals. That setup favors buying quality AI-hardware names on weakness rather than joining the stampede out.

## Idea

The AI selloff is hitting every chip and cloud stock, but the demand side of the story has not changed: ByteDance is still negotiating a large data-center chip order with MediaTek, and analysts are calling a bottom in AI infrastructure names like CoreWeave. When prices fall but customers keep signing big deals, the selling is more about market mood than broken fundamentals. That setup favors buying quality AI-hardware names on weakness rather than joining the stampede out.

## Advanced Analysis

### Verdict: The Demand Story Is Real, But the Signal Hasn't Fired — Wait

\*\*Verdict: the thesis deserves a spot on the watch list, not the buy list — wait for the reversal to confirm.\*\* The strongest point for the idea is Micron's quarter ended September 3, 2026: revenue of $54.2B up 30.8% quarter over quarter, net income of $37.7B, a 69.5% net margin, and free cash flow of $32.9B, up 87.1% — hard to square with broken demand. The strongest point against is that every entry condition is far from met (RSI at 19.7 on MU and 17.1 on CRWV versus a 30 trigger; MU trades $99.85 below its 21-day EMA at $1,013.98, CRWV $18.35 below at $83.44), while the provisional September 30, 2026 ownership filings show roughly $231.1M of net open-market insider selling at Micron — the people running this cycle selling into it. CoreWeave's fundamentals also cut against the 'mood not fundamentals' framing: gross margin fell 2.1 points to 65.5%, net margin worsened to negative 35.6%, quarterly free cash flow was negative $4.7B, and shares outstanding grew 5.9% to 532M, against $46.0B of liabilities. What would flip this to a buy: confirmation of the entry conditions (an RSI cross back above 30 plus a close above the 21-day EMA on both names) alongside a flip to net insider buying in the next ownership filing after the September 30 report period. Until then, this is a setup waiting for its own rules, not an active signal — and note that the third name in the thesis, MediaTek (2454.TW), has no supported fundamentals coverage here, and no robust parameter setup was established because CRWV daily data has an unfilled gap. The conviction scores below reflect that split: solid thesis support, weak trigger proximity.

#### Conviction score breakdown

Composite score computed by the server from the applicable evidence-tier dimensions.

| Measure | Value |
| --- | ---: |
| Thesis support | 65/100 |
| Trade readiness | 35/100 |
| Risk quality | 40/100 |
| Trigger proximity | 20/100 |
| Fundamentals trend | 55/100 |
| Score | 43/100 |
| Composite Score | 43/100 |
| Evidence Tier | rules\_not\_triggered |

#### Decision scenarios

Bull, base, and bear cases synthesized from the cited evidence tier. Likelihoods are rounded evidence-weighted judgments, not statistically calibrated forecasts.

| Measure | Value |
| --- | ---: |
| Evidence Tier | rules\_not\_triggered |

### Trade now: the entry is a recovery trigger, not a knife-catch

Nothing is live yet, and the setup is designed so you should not want it to be. The idea argues the AI selloff reflects sentiment, not broken demand, so the entry waits for confirmation. On each name, the rules require an RSI (14) cross back above 30, a close above the 21-day EMA, trend strength (ADX 14) above 20, and a low touching the nearest support level. Today none of those conditions is close on either ticker.

\*\*CRWV\*\* is the nearer candidate. At $65.09, it sits essentially on its first support at $65.11, so the support-test condition is effectively satisfied. But RSI (14) is 17.1 — it needs to turn and cross above 30, not just stabilize — the close is $18.35 below the 21-day EMA at $83.44, and ADX (14) is 13.9 versus the 20 threshold. That is a washed-out market waiting for a reversal, not one mid-turn.

\*\*MU\*\* is further away despite stronger fundamentals (quarterly revenue of $54.2B, up 30.8% quarter over quarter, with a 69.5% net margin). At $914.13 it is $99.85 below its 21-day EMA at $1,013.98, RSI (14) is 19.7 versus the 30 trigger, ADX (14) is 15.8 versus 20, and first support sits at $911.33. Support is being tested, but the momentum reclaim has not started.

The risk math, once triggered, is mechanical: a hard stop at a 2.6% loss on the position and a take-profit at a 5.2% gain, an effective reward-to-risk of about 2-to-1 — with a backstop exit if price closes back above the second support level (MU at $901.3, CRWV at $68.51) or if a position is held for 90 daily bars. "Wait" here means one concrete thing: do not buy the support touch. Wait for the RSI cross above 30 and the close back above the 21-day EMA, then size the position to a fixed risk of roughly 2.6% of capital with no single name above 25% of the portfolio. One scope note: the parameter study for this setup could not be completed because daily data for CRWV has an unfilled gap, so no robust parameter setup was established — the published thresholds stand as written.

#### CRWV price and trigger map

Uses the idea timeframe and keeps price levels on the price axis.

| Measure | Value |
| --- | ---: |
| Ticker | CRWV |
| Timeframe | 1d |

#### MU price and trigger map

Uses the idea timeframe and keeps price levels on the price axis.

| Measure | Value |
| --- | ---: |
| Ticker | MU |
| Timeframe | 1d |

### Why the Bull Case Still Has Support

The idea's core claim is that the AI selloff is a mood problem, not a demand problem, and Micron's latest reported quarter is hard to reconcile with broken fundamentals. For the fiscal quarter ended September 3, 2026, Micron reported revenue of $54.2B, up 30.8% quarter over quarter, with net income of $37.7B and net margin of 69.5% — a full percentage point above the prior quarter. Free cash flow hit $32.9B in the quarter, up 87.1% from $17.6B the quarter before. Against a peer set of 805–899 Information Technology names, Micron sits in the 95th percentile on gross margin (86.8%), the 99.8th percentile on operating margin (80.7%), and the 99.5th percentile on free cash flow. That is not what a business in decline looks like; the per the Reuters report on October 15, 2026, the ByteDance–MediaTek data-center chip negotiation is a live, unsigned but active demand conversation, which is exactly the kind of customer behavior the thesis leans on. CoreWeave — the name the thesis specifically flags as bottom-called — also shows demand growth on the top line even as the selloff bites. Revenue reached $2.08B in the quarter ended March 31, 2026, up from $1.57B the prior quarter and $982M a year earlier, while operating cash flow was positive at $3.0B for the quarter. Per the CNBC piece from October 15, 2026, Deutsche Bank is explicitly calling CoreWeave's share weakness a buy-the-dip opportunity, which aligns with the idea's 'buy quality on weakness' framing rather than contradicting it. A second argument for the thesis is the divergence itself. Micron is generating record cash while the market reprices AI hardware, and CoreWeave is still growing revenue over 100% year over year on a small base. The thesis's value framing — that falling prices with intact customer commitments create an asymmetric entry — is at least internally…

#### CRWV Revenue

Revenue trend from CommonQuant fundamentals/XBRL data; +13027% from first to latest point.

| Measure | Value |
| --- | ---: |
| 2022-12-31 | $15830000 |
| 2023-12-31 | $229000000 |
| 2024-12-31 | $1915000000 |
| 2025-03-31 | $982000000 |
| 2025-06-30 | $1212788000 |
| 2025-09-30 | $1364676000 |
| 2025-12-31 | $5131000000 |
| 2025-12-31 | $1571904000 |
| 2026-03-31 | $2078000000 |
| Latest Value | $2078000000 |
| Change Pct | $13026.974099810484 |
| Ticker | CRWV |
| Timeframe | reported periods |

#### CRWV Gross margin

Gross margin trend from CommonQuant fundamentals/XBRL data; +179.8% from first to latest point.

| Measure | Value |
| --- | ---: |
| 2022-12-31 | \0.23423878711307644% |
| 2023-12-31 | \0.6986899563318777% |
| 2024-12-31 | \0.7425587467362925% |
| 2025-03-31 | \0.7331975560081466% |
| 2025-06-30 | \0.7421915454308584% |
| 2025-09-30 | \0.7297351166137603% |
| 2025-12-31 | \0.7168193334632625% |
| 2025-12-31 | \0.6761157169903506% |
| 2026-03-31 | \0.6554379210779595% |
| Latest Value | \0.6554379210779595% |
| Change Pct | \179.8161351311785% |
| Ticker | CRWV |
| Timeframe | reported periods |

#### MU sector percentile check

Ranks MU against 899 companies in its sector using CommonQuant fundamentals.

| Measure | Value |
| --- | ---: |
| Operating margin | \99.77753058954394th percentile |
| Free cash flow | \99.51980792316928th percentile |
| Revenue growth (YoY) | \96.7741935483871th percentile |
| Gross margin | \95.1552795031056th percentile |
| Ticker | MU |
| Sector | Information Technology |
| Peer Count | 899 |

### Scores

- **Conviction score breakdown:** 43
- **Thesis support:** 65
- **Trade readiness:** 35
- **Risk quality:** 40
- **Trigger proximity:** 20
- **Fundamentals trend:** 55

### Watch items

- **CRWV — RSI (14) cross above 30**
- **CRWV — Close above 21-day EMA**
- **CRWV — ADX (14)**
- **CRWV — Price vs first support**
- **MU — RSI (14) cross above 30**
- **MU — Close above 21-day EMA**
- **MU — ADX (14)**
- **MU — Insider net open-market activity**
- **CRWV — Gross margin (Q1 2026)**
- **CRWV — Shares outstanding**

## Key details

- Symbols: \2454.TW, CRWV, MU
- Timeframes: 1d
- Tags: \#canonical-demand, \#cluster-version:1, \#direction:bullish, \#entity-kind:instrument, \#entity:2454.TW, \#entity:CRWV, \#entity:MU, \#horizon:unspecified, \#intent:research, \#symbol:2454.TW, \#symbol:CRWV, \#symbol:MU

## Community

- Upvotes: 1
- Views: 3
- Copies: 0
- Cosigns: 0

## News sources

- [ByteDance nears chip supply deal with MediaTek for data centres, Bloomberg News reports - Reuters](https://news.google.com/rss/articles/CBMivgFBVV95cUxQSzBSa0l3cFM5aWVOV3JMUUNqT2JVbTREN0puV0V0TW1zU0Z3cXZuQ29KR3pQelFyVU5CeTJOMHBfU2lNN3Y0SmhFSkpSUlI4NGtRdTA4M2kwSUZpQUdkRlpmcTFISlA3b0QzVWoyUi1zcUh5ZEtScHBCRGZkU2VnLXYwcjNHX01VNnpKS0J5YkFGRU9IQlV4VDI0U250NjhfQWhtWW12Tm9SZHpNOU9hSzVCa1BFeGpjWW41czV3?oc=5) — Reuters
- [CoreWeave shares are sinking. Buy the dip, Deutsche Bank says](https://www.cnbc.com/2026/10/15/ukraine-russia-energy-ceasefire-zelenskyy.html) — CNBC

## Related

- [\2454.TW trade ideas](https://commonquant.ai/markets/2454.tw)
- [CRWV trade ideas](https://commonquant.ai/markets/crwv)
- [MU trade ideas](https://commonquant.ai/markets/mu)
- [Latest market news](https://commonquant.ai/news)

## About CommonQuant Research

Ideas and Advanced Analysis are generated with fresh, LLM-selected news headlines and grounded in SEC XBRL fundamentals and peer percentiles. Strategies built from these ideas are automatically backtested (over a timeframe-dependent historical window with walk-forward tuning) and stress-tested before they can go live, then monitored against the news hourly while they run.