Semiconductors will reign over the next several years, with foundries, memory, photonics, photonics foundries and OSATs the areas to be long. x.com  x.com 

Applied Materials' larger stake in BE Semiconductor points to long-term hybrid-bonding potential despite delayed adoption in HBM. Applied agreed to raise its BESI holding from 9% to about 20%; HBM adoption timelines may not reflect the prospects for the technology as a whole. x.com  x.com 

Semiconductors are set up for a violent, DRAM-style bust because demand again hinges on a few big customers and one theme, an AI training and benchmarking phase that will not last. NVIDIA is hurt badly if even one customer pulls back. Unlike Cisco's fungible parts, it funds custom TSMC lines, with $182 billion of purchase commitments exceeding its annual operating cash flow. Microsoft's receivables with NVIDIA jumped to about $12.2 billion, suggesting it pulled forward chips it is warehousing. … Show more

Semiconductors will be much higher in 12 months if the market's real doubt is whether this cycle lasts another year. The cycle feels like the bottom of the third inning while the market prices a seventh-inning stretch, which is why so many chip names look like must-owns at these valuations. In order of preference: Nvidia, Credo, Astera Labs, Lumentum, Broadcom, TSMC, Coherent, Marvell, then SK hynix and Micron. x.com 

Chip-equipment makers look cheap given supply-chain shortages and demand visible toward 2029-30. At Semicon West, no one raised demand concerns; suppliers' suppliers lacked capacity, while several high-quality names trade below the S&P multiple. Per Lynx's KC, customers are asking about delivery three to four years out. For Lam Research's report, bottlenecks, pricing and visibility beyond 2027 matter more than demand. x.com  x.com 

The semiconductor trade looks fragile because it now hinges on AI labs' growth. Semis sold off when Anthropic's month-over-month ARR growth decelerated, putting the market's fragility on full display, and that may be a sign of things to come. x.com 

Kulicke and Soffa is mispriced because JEDEC's relaxation to 775μm extends thermocompression bonding into HBM4 and HBM5, which is KLIC's field. Its fluxless TCB tools, which with ASMPT's aim to delay hybrid bonding adoption, are a structural tailwind the market reads as a delay risk. The stock has been a chronic disappointer for about three years on wirebonder cyclicality, and the dominant TCB narrative belongs to BESI and ASMPT. x.com 

Semiconductor demand for 2027 is locked in, with order books already done for next year, so arguments about semis, including zero-sum debates, miss the point. @danielnewmanUV notes Nvidia's deal with Amazon to deploy 2 million more GPUs plus Vera shows supply is constrained everywhere and AI demand is insatiable; it won't be all Nvidia or all Trainium. x.com 

The semiconductor playbook that worked in 2025 and the first half of 2026 will likely fail from here. The era of bottlenecks, when five years of fundamentals were priced in within five weeks, is over, and so far a V-shaped recovery in AI and semis is not taking place. The cooling is probably healthy and calls for patient stock picking across subniches, with valuations hopefully mattering more. If the old regime returns, a more aggressive stance would fit again. x.com 

Slow capacity ramps and long-term deals should tighten memory supply next year and support earnings. Fundamentals look stronger than Q1: Micron expects 2027-28 tighter than 2026, with over 75% of FY27 output committed. Samsung, SK hynix and SanDisk also have multi-year deals. AI-trade positioning and sentiment remain risks; CXMT faces US restrictions and qualification barriers in HBM and hyperscaler RDIMM. A midterm/Christmas rally into 2027 remains the prediction. x.com  x.com