#美股AI概念股普涨

---

1️⃣ Recent purchase of US stocks with AI concepts and trading ideas on Gate

Recently, in Gate's US stock trading area, three main targets were focused on: Micron (MU), Intel (INTC), and a small amount of Nvidia (NVDA). The specific ideas are as follows:

Micron (MU) —— The strongest in storage logic

· Buying point consideration: Micron surged nearly 10% in a single day on June 8, but had been correcting for three consecutive weeks prior. Observing that HBM capacity remains in short supply, Micron's HBM capacity for 2026 has been fully booked. On June 8, the NASDAQ rebounded nearly 6%, with MU leading the storage sector, confirming a short-term sentiment turning point.
· Operation method: Use USDT on Gate to directly buy MU shares, building positions in two steps——the first within 2 hours after the rebound starts, buying a small amount, and the second after a slight intraday pullback, adding more.
· Risk control setup: Set a stop-loss at 5% below the low point of June 5, expecting to hold until the end of June or the rebalancing window.

Intel (INTC) —— Order-driven reversal play

· Logic: Google’s 3 million AI chip orders will be delivered by 2028, with actual performance realization still distant, but the market chose to "believe first, see later." Intel previously experienced severe overselling, with a large volume of orders catalyzing at low levels, making short-term odds attractive.
· Operation: Participate with small positions on Gate, adopting a quick in and out approach. When Intel rose 11% in a day without chasing, observe the trading volume the next day; if it does not significantly shrink, then enter at a 3% pullback.
· Note: Do not view it as a long-term hold, more as an event-driven trading.

Nvidia (NVDA) —— Follow-up buying strategy

· NVDA only rose 1.73% that day, underperforming the sector, but as the absolute leader in AI computing power, it exists as a "core holding" in Gate’s US stock portfolio. Core idea: do not pursue short-term excess returns from NVDA, but use Gate’s fragmented trading feature to invest small amounts weekly in USDT, smoothing costs.

---

2️⃣ Analysis of AI concept stocks' future trend under the strong alliance of giants

Short-term (1-3 months): Beware of "order false fire," inevitable divergence

After the announcement of 3 million AI chip orders from Intel, the stock surged 11%, but note: this order will only be delivered in 2028, with almost zero impact on Intel’s revenue and profit in the next two years. This kind of "distant water" to quench "near thirst" rally is essentially an emotional release in a liquidity-rich environment.

Possible divergence paths in the coming weeks:

· Targets truly benefiting from current AI capital expenditure (Nvidia, Broadcom, Micron, TSMC) are expected to maintain strength, especially on the storage side, where HBM capacity shortages may last until at least 2027.
· Targets relying on long-term orders or riding on concepts, once the market adjusts (e.g., CPI/PPI in June exceeding expectations causing rate hike fears), will see sharp corrections.

Mid-term (6-12 months): Energy and connectivity become "hidden champions"

Current discussion mainly focuses on chips themselves, but power bottlenecks in AI data centers and connectivity bottlenecks are becoming more critical than chips:

· Power side: Traditional grid expansion takes 5-7 years, gas turbine orders are scheduled until 2029, GE Vernova’s backlog reaches 150 billion USD. Meanwhile, Bloom Energy’s fuel cells for power supply only take 55 days to sign, with Q1 revenue up 130%. These targets can be watched on Gate: GE Vernova (GEV), Bloom Energy (BE), currently with much lower capital attention than chip stocks, presenting an expectation gap.
· Connectivity side: Lumentum and other optical module manufacturers have gained over 145% this year, but their valuations are still below those of chip design companies, with room for catch-up.

Long-term (more than 1 year): Signals from Cathie Wood’s operations

Wood significantly reduced TSMC holdings and took profits on AMD in the first five months of 2026, while building positions in Cerebras (AI inference chips at wafer level). This conveys a key judgment: AI computing power demand is shifting from large-scale model training to high-frequency, low-cost inference.

This implies:

· The "barbaric growth" phase of training chips may be over, with growth slowing down.
· Inference chips, edge computing, and low-power solutions will become the next main battleground.
· For individual investors, it is still reasonable to heavily hold leading training chips (like Nvidia), but attention should start to shift to early opportunities in inference, using small positions on platforms like Gate in segments not yet fully priced.

Overall judgment

Optimistic about the medium- and long-term trend of AI concept stocks, but in the next 6 months, more inclined towards segments "beyond computing power"——storage (Micron, SanDisk), power (GEV, BE), optical connectivity (Lumentum), which may have greater potential than the already surged chip design leaders. Meanwhile, closely monitor May CPI data; if inflation exceeds expectations, tech stocks will face valuation pressures in the short term, making it a good time for phased deployment rather than panic exit.

---

The above is a personal investment idea sharing, not investment advice. Gate’s USDT trading feature for US stocks provides a convenient cross-market allocation tool, but leverage, options, and other derivatives carry higher risks. Ordinary users are advised to focus mainly on spot trading.