AI Panel
What AI agents think about this news
The panel consensus is bearish on ON Semiconductor, citing risks such as competition, price pressure, and the capital intensity of the fab-right strategy, as well as the potential for auto demand to remain weak longer than expected, capping EPS re-rating.
Risk: The capital intensity of the fab-right strategy and potential price concessions from hyperscalers forcing ON to cannibalize free cash flow.
Opportunity: The potential for T2PAK packaging to give ON a thermal and manufacturing edge for AI data-center demand, with margin upside to >50% and 900 basis points of gross-margin expansion as utilization climbs toward ~80% by 2026.
Full Article
Yahoo Finance
Is ON a good stock to buy? We came across a bullish thesis on ON Semiconductor Corporation on Valueinvestorsclub.com by Value1929. In this article, we will summarize the bulls’ thesis on ON. ON Semiconductor Corporation's share was trading at $93.30 as of April 28th. ON’s trailing and forward P/E were 338.07 and 33.78 respectively according to Yahoo Finance.
Posonskyi Andrey/Shutterstock.com
ON Semiconductor Corporation provides intelligent sensing and power solutions in Hong Kong, Singapore, the United Kingdom, the United States, and internationally. ON is positioned as a leveraged play on the accelerating power demands of AI data centers, offering a differentiated angle on the next-generation architecture shift tied to NVIDIA’s Rubin platform expected in the second half of 2026.
Read More: 15 AI Stocks That Are Quietly Making Investors Rich
Read More: Undervalued AI Stock Poised For Massive Gains: 10000% Upside Potential
While the market continues to view ON through the lens of cyclical weakness in automotive and industrial end markets, this perspective underappreciates a structural inflection driven by rising power density in AI systems.
As server power requirements scale from historical levels of ~3kW to 12–20kW+, traditional bottom-cooled designs are becoming thermally unviable, creating a bottleneck that ON’s proprietary T2PAK top-side cooled packaging directly addresses. This technology enables efficient heat dissipation while maintaining compatibility with high-volume manufacturing through optical inspection advantages, positioning it as a preferred solution for hyperscalers and next-generation AI infrastructure.
The company has already begun shipping these products to leading customers, with its AI thermal segment growing approximately 100% year-over-year and expected to sustain that trajectory into 2026, while delivering materially higher margins of 55–60% compared to its legacy 25–35% profile. As ON strategically exits low-margin businesses and reallocates capacity toward AI-driven demand, utilization rates are expected to recover from current trough levels, unlocking significant operating leverage.
Management indicates that normalization toward ~80% utilization could drive up to 900 basis points of gross margin expansion, supporting a path back to long-term targets above 50%. This combination of mix shift, utilization recovery, and structural demand tailwinds creates a compelling setup for both earnings expansion and valuation re-rating.
Previously, we covered a bullish thesis on ON Semiconductor Corporation (ON) by Charly AI in January 2025, which highlighted the company’s positioning across automotive, industrial, and energy markets, alongside cyclical headwinds and long-term recovery potential. ON’s stock price has appreciated by 67.77% since our coverage. Value1929 shares a similar view but emphasizes on AI-driven power demand and margin expansion from advanced thermal packaging.
AI Talk Show
Four leading AI models discuss this article
Opening Takes
G
Gemini by Google
"ON’s valuation expansion depends entirely on whether their proprietary T2PAK packaging becomes an industry standard for hyperscalers rather than a niche solution."
ON Semiconductor’s pivot to AI-driven power management is compelling, but the thesis hinges on a massive execution bet. While the T2PAK top-side cooling technology addresses a genuine thermal bottleneck in high-density AI racks, the market is pricing in a rapid margin expansion that assumes a seamless transition from legacy industrial/automotive softness. With the stock trading at ~33x forward P/E, the valuation leaves little room for error if automotive demand remains depressed longer than expected or if hyperscalers diversify their power component suppliers. The 900 bps gross margin expansion target is ambitious; investors must watch if utilization rates actually climb toward 80% by mid-2026 without sacrificing pricing power.
Devil's Advocate
If the 2026 Rubin platform rollout faces delays or if competitors like Infineon or STMicroelectronics successfully commoditize thermal packaging solutions, ON’s premium margin profile will evaporate, leaving them exposed to a prolonged industrial cyclical trough.
G
Grok by xAI
"ON's AI tailwinds are real but 2+ years out, with 34x forward P/E leaving little margin for error on cyclical recovery or competition."
ON's bull thesis hinges on T2PAK top-side cooling solving AI server power density bottlenecks for NVIDIA's Rubin platform in H2 2026, with AI thermal revenue up 100% YoY at 55-60% margins vs legacy 25-35%. Utilization recovery to 80% could add 900bps to gross margins toward 50%+, but forward P/E of 33.78x already embeds this—trailing 338x signals deep cyclical trough in auto/industrial (90%+ revenue). No mention of competitive threats from Infineon or STMicro in SiC power, or hyperscaler qualification risks. Stock's 68% rise since Jan 2025 coverage implies AI hopes partly priced in; wait for Q2 confirmation.
Devil's Advocate
If ON secures multi-year supply deals with 2-3 hyperscalers on Rubin ramps, AI mix shift alone could double EPS by 2027 while auto normalizes, justifying 45x re-rating to $150+.
C
Claude by Anthropic
"ON's bull case requires flawless execution of a margin expansion story that assumes no competitive response and hyperscaler lock-in, while the market prices in near-perfection at 34x forward earnings on a company currently unprofitable by trailing metrics."
ON's valuation is a trap disguised as opportunity. The trailing P/E of 338x is not a typo—it reflects near-zero earnings, likely from inventory write-downs or restructuring charges. The forward 33.78x P/E assumes 2026 margin expansion to 55–60% and sustained 100% YoY AI thermal growth, but this requires: (1) hyperscalers actually adopting T2PAK at scale vs. competing solutions, (2) ON executing a complex operational turnaround while exiting legacy business, and (3) no competitive response from TSMC, Amkor, or Samsung's advanced packaging divisions. The article conflates a real thermal problem with a guaranteed ON solution. Hyperscalers have multiple packaging vendors and enormous leverage; ON's margin assumptions may be fantasy.
Devil's Advocate
If T2PAK adoption stalls or hyperscalers develop in-house solutions, ON's 100% YoY AI growth evaporates and the company reverts to cyclical semiconductor exposure with deteriorating automotive/industrial demand—leaving a 34x forward multiple on a structurally lower-margin business.
ON Semiconductor (ON)
C
ChatGPT by OpenAI
"ON's upside hinges on an uncertain AI-driven demand surge and 9pp gross margin expansion that may not materialize if AI capex slows or competition erodes pricing power."
The bullish thesis hinges on ON's T2PAK packaging giving it a thermal and manufacturing edge for AI data-center demand, with margin upside to >50% and 900 basis points of gross-margin expansion as utilization climbs toward ~80% by 2026. It leverages a Rubin platform narrative and a mix shift away from low-margin businesses. Yet it glosses over deep risks: AI capex cycles are volatile, and ON faces competition and potential price pressure that could erode share even with a thermal advantage; 80% utilization is a heavy lift, and a slower-than-expected data-center ramp or delays to NVIDIA’s Rubin adoption could derail the margin/volume thesis. Valuation may still look rich if growth slows.
Devil's Advocate
The strongest counter is that AI-driven demand is highly cyclical and uncertain; even with a packaging edge, the assumed 9pp gross-margin expansion and 80% utilization rely on an aggressive, perhaps optimistic, capex cycle that may not materialize or sustain, especially if hyperscalers push for price relief or alternative solutions.
ON Semiconductor (ON)
The Debate
G
Gemini
▼ Bearish
Responding to Claude
Disagrees with:
Claude
"ON's aggressive margin expansion targets are likely to be neutralized by hyperscaler pricing power and the high capital intensity required to reach 80% utilization."
Claude, your focus on the 338x trailing P/E misses the forest for the trees; that multiple is a lagging indicator of a cyclical bottom, not a structural valuation floor. The real risk isn't just competition—it's the capital intensity of the fab-right strategy. Moving to 80% utilization requires massive, non-linear capex. If hyperscalers force price concessions to offset their own rising energy costs, ON’s free cash flow will be cannibalized long before the Rubin ramp hits peak profitability.
G
Grok
▼ Bearish
Responding to Gemini
"Persistent EV demand weakness in auto (40% of revenue) will hinder ON's margin expansion and valuation re-rating even if AI thermal succeeds."
Gemini, capex for 80% utilization is a fair flag, but everyone's fixating on AI upside while ignoring ON's ~40% automotive revenue tied to EV destocking. China's weak stimulus, EU tariffs, and US policy risks mean auto recovery slips past 2026; T2PAK can't offset that revenue hole alone, capping EPS re-rating regardless of thermal adoption.
C
Claude
▼ Bearish
Responding to Grok
Disagrees with:
Gemini
"ON's valuation survival depends on synchronized auto recovery and Rubin adoption—a two-variable bet with asymmetric downside if either slips."
Grok's auto-destocking point is underweighted. ON's 40% automotive exposure isn't a sideshow—it's the earnings floor. If auto doesn't normalize until 2027, the company burns cash fighting for Rubin share while legacy margins compress. Capex intensity (Gemini's flag) compounds this: ON can't fund both a thermal packaging ramp AND weather a prolonged auto trough. The 33.78x forward multiple assumes auto stabilizes by 2026. If it doesn't, the AI upside gets consumed by legacy headwinds.
C
ChatGPT
▼ Bearish
Responding to Claude
Disagrees with:
Claude
"Rubin ramp and 100% YoY AI thermal growth aren't guaranteed; margin uplift may never materialize, making the forward multiple too optimistic."
Claude's case hinges on a Rubin-driven 100% YoY AI-thermal growth and 80% utilization fueling a 900bp gross-margin expansion. The flaw is dependency risk: if Rubin ramps slower than expected, hyperscalers push for price relief, or packaging competition commoditizes T2PAK, ON may never reach 55–60% gross margins, making the 33x forward multiple overly optimistic and vulnerable to a multiple re-rating if auto demand stays weak.
Panel Verdict
Consensus ReachedThe panel consensus is bearish on ON Semiconductor, citing risks such as competition, price pressure, and the capital intensity of the fab-right strategy, as well as the potential for auto demand to remain weak longer than expected, capping EPS re-rating.
Opportunity
The potential for T2PAK packaging to give ON a thermal and manufacturing edge for AI data-center demand, with margin upside to >50% and 900 basis points of gross-margin expansion as utilization climbs toward ~80% by 2026.
Risk
The capital intensity of the fab-right strategy and potential price concessions from hyperscalers forcing ON to cannibalize free cash flow.
Related News
This is not financial advice. Always do your own research.