Shares of Nvidia (NASDAQ: NVDA) may be flashing a warning despite hitting fresh highs, as a key momentum indicator suggests the rally could be overstretched.
This technical backdrop contrasts with the stock’s strong recent performance, with Nvidia closing at a record $208.27 on Friday, marking its first all-time closing high since October 2025.
The rally briefly pushed its market capitalization back above $5 trillion during intraday trading.
However, the surge has lifted the stock’s 14-day Relative Strength Index (RSI) to 71.50, pushing it into overbought territory and signaling an increased risk of a near-term pullback even as prices continue to rise.
On a comparative basis, the semiconductor company ranks worse than 82.26% of its peers on this metric, with the industry median notably lower at 59.265.
Even so, the RSI signal introduces a layer of caution. The stock has gained roughly 20% over the past month, a pace that typically invites consolidation as traders take profits and momentum cools.
While overbought conditions do not guarantee a reversal, they often precede periods of heightened volatility.
Such a gap indicates that Nvidia’s gains have outpaced much of the broader sector, a pattern often associated with short-term exhaustion.
Nvidia’s historic run
Notably, the Nvidia stock surge has been driven largely by a broader rebound in semiconductor stocks.
This was partly triggered by a strong earnings report from Intel (NASDAQ: INTC), which lifted sentiment across the chip sector and sparked gains in companies such as Advanced Micro Devices (NASDAQ: AMD) and Qualcomm (NASDAQ: QCOM).
At the same time, optimism around artificial intelligence infrastructure and rising data center demand has reinforced Nvidia’s central role in the AI ecosystem.
Investor enthusiasm has also been supported by expectations of continued growth tied to hyperscaler spending and next-generation AI platforms.
This is largely centered on the Vera Rubin platform, unveiled at CES 2026 and expanded at GTC 2026, which introduces a full-stack AI architecture promising up to 10x lower inference costs and higher performance.
Shipments began in early 2026, with large-scale cloud deployments expected later this year.
CEO Jensen Huang has projected that combined sales of Blackwell and Vera Rubin chips could reach $1 trillion by 2027, highlighting the scale of AI infrastructure growth.
