While the market applauds giant chipmakers, quiet fortunes are compounding inside the second-wave stocks behind them; power delivery, optical connectivity, data storage, fiber, and next-generation brokerage models decide who really banks the AI spending boom, and this analysis puts five concrete candidates with entry levels on the table.
At the heart of the Vicor thesis sits a simple physics problem: ever more powerful processors need huge current at low voltage, and current traveling across a circuit board wastes energy as heat; the company's vertical power delivery technology shortens that path by feeding power from beneath the processor, which becomes decisive as data center architectures grow more complex.
The numbers back the story; the stock had gained about 366 percent year to date as of September 22, with revenue growth of 36 percent for 2026, 56 percent for 2027, and 23 percent for 2028, while gross margin is seen between 57 and 64 percent and the EBITDA margin should climb from 23 toward 44 percent; according to everyticker.com, the company collected 30.4 million dollars in licensing revenue in the second quarter, about 21 percent of total revenue, and with a 52 times forward multiple the 280 to 300 dollar band with an RSI of 72 frames the entry.
Vicor: the power bottleneck and the licensing engine
At MaxLinear the story is written in optical connectivity; the Keystone platform covers 400G and 800G lines while the Rushmore 1.6T PAM4 DSP moves through customer qualification, and the shift to 200G lane speeds offers real share gains in AI data center markets; according to convergedigest.com, pairing Rushmore with the Washington TIA implies more than 100 dollars of semiconductor content per module, which lays the optical leverage bare.
The company grew 55 percent in the second quarter, with revenue growth of 60 percent penciled in for 2026 and 32 percent for 2027, while the EBITDA margin is seen rising from around 3 percent toward 25 and then 34 percent, and trading at 30 times EBITDA shows how powerful the operating leverage is; the 90 to 100 dollar band with an RSI of 65 and upward torque points to a disciplined chart setup.
MaxLinear: leverage from the optical transition
On the Everpure front, the company long known as Pure Storage, the FlashBlade EXA platform targets enterprise AI storage; the rename recorded in a February 2026 SEC filing mirrors the ambition of selling flash storage to hyperscalers at hyperscale, and fast data access for AI workloads forms the technical backbone of the thesis.
At its analyst day the company guided fiscal 2028 revenue to 7 billion dollars, clearly above the 6.5 billion consensus, and the shares closed at a record 121.88 dollars on September 24 as the index's top performer; according to tickergrove.com, fiscal 2027 growth of 38 percent and fiscal 2028 growth of 37 percent with a 70 percent gross margin are targeted, and with a 25 times EBITDA multiple and an RSI of 74 the sub-120 dollar zone is watched as the entry area.
Everpure and Corning: storage meets fiber
The Corning thesis compresses into one line: AI data centers need roughly ten times more fiber than traditional builds, and traffic between dense compute clusters turns glass fiber demand into a structural growth line; this fiber intensity dynamic makes the company a direct beneficiary of the infrastructure spending wave.
That picture hardened with a giant deal on September 29, 2026; according to reuters.com, AT&T signed a multi-year fiber supply contract worth more than 3 billion dollars with Corning, and after an 80 percent climb in 2026 the 160 to 170 dollar band is watched for entry while 200 and then 250 dollars are discussed as targets.
Robinhood and the big picture
At Robinhood the fastest-growing business is no longer stock commissions; according to fool.com, prediction-market revenue reached 156 million dollars in the second quarter, up more than tenfold from a year earlier, outrunning 29 percent growth in options and 95 percent in equities while dwarfing a 38 percent drop in crypto, and the line now makes up about one fifth of transaction revenue.
Crypto offers a second catalyst; according to galaxy.com, the 50-week average acts as a resistance ceiling and the 200-week average as a floor for Bitcoin, and the early-September reclaim of the 50-day average plus the fifth halving cycle expected in spring 2028, combined with Vlad Tenev's everything-exchange vision, completes the story.
The macro frame carries all these theses; according to goldmansachs.com, global AI investment should exceed 1 trillion dollars in 2026, with hyperscalers spending about 800 billion dollars on a path toward 1.5 trillion in 2027, and optical plus inference-side demand revives the CPU side too, spreading the spending wave across every infrastructure layer.
Discipline closes the loop; charts that hold a rising 200-day average, turn a short-term dip into a short-term rebound, and signal a resumption of the long-term uptrend stand out as classic fourth-quarter candidates, and scaling in gradually without chasing stretched RSI readings remains the simplest way to manage risk in this fast-moving group.
Key moments
AI commentary
"The narrative is strong and backed by numbers, yet multiples are rich and RSI readings run hot. Taking the theses seriously while staging entries gradually looks like the soundest posture."
AI assessment
The strongest counter-argument comes from valuation; a 52 times forward multiple, a triple-digit yearly gain, and RSI readings above 70 suggest the good news is largely priced in, and a single fab delay or a slowdown in licensing pace could compress those multiples fast.
Gaps remain as well; customer qualification calendars, supply-chain risks, and regulatory uncertainty around prediction markets sit in the background, and on the Robinhood side swings in consumer confidence plus the crypto cycle make the fastest-growing revenue line the most fragile one.
The speaker's possible interest should not be ignored; a market commentator who loves trend-resumption setups will naturally spotlight high-momentum names, and while the entry bands and RSI thresholds are practical, they are frameworks to adjust to personal risk appetite rather than guarantees.
The practical takeaway for readers is clear; positions should be built in stages, a stop level should be written down for every name before buying, and the fourth-quarter expectation should never be a standalone reason to buy without macro confirmation, because in this fast group discipline beats prediction.
Sources
8 links; 1 of them also cited by 4 other stories. Stories sharing a link do not confirm each other; a source's origin is not inferred from how often it is cited.
- @youtube.com YouTube — Being Exponential with Luke Lango
- @everyticker.com EveryTicker
- @convergedigest.com Converge Digest
- @tickergrove.com TickerGrove
- @reuters.com Reuters
- @fool.com Motley Fool
- @goldmansachs.com Goldman Sachs
Also cited by: Memory rally: why chip stocks soared in 2026 and the cheap-looking trap · Everything Screams Crash Yet Stocks Keep Climbing Higher Anyway · Everyone Hates AI Right Now: Four Stocks That Stay Bulletproof · The $7.6 Trillion AI Cake: Five Layers, Two Chip Giants, and My Map
- @galaxy.com Galaxy
ai · stocks · vicr · maxlinear · robinhood