Micron Technology will supply high-bandwidth memory and storage to Anthropic and invest in its $65B Series H, deepening hardware-software integration for cloud-scale AI. The partnership will co-optimize memory for Claude workloads, reducing bottlenecks as the $965B-valued company races toward IPO. For cloud and SaaS providers, this signals a new phase where memory performance directly shapes AI service reliability and cost.
Source: telecom.economictimes.indiatimes.com · 933thedrive.com
Apple is pouring $30 billion into Broadcom to manufacture more than 15 billion chips, including custom ASICs for artificial intelligence workloads. The decade-long agreement ensures Apple’s AI hardware roadmap stays proprietary and vertically integrated, free from GPU shortages.
Apple’s largest-ever domestic manufacturing commitment will direct over $30 billion to Broadcom, securing 15 billion U.S.-made chips and expanding a Colorado fab. The deal marks a seismic shift toward reshoring critical semiconductor production, reducing dependency on Asian suppliers.
Source: proactiveinvestors.com · CNBC
DeepSeek's development of in-house AI inference chips signals a potential shift in semiconductor procurement, reducing reliance on Nvidia and Huawei while aligning with China's domestic chip push amid US export restrictions.
DeepSeek, China's AI champion, is reportedly developing its own inference chips—a high-stakes expansion that could redefine its valuation and challenge VC-backed AI hardware startups. The move signals a new phase in the AI chip race.
Chinese AI leader DeepSeek is developing its own inference chips, mirroring moves by OpenAI, to reduce Nvidia reliance. This shift underscores the growing importance of inference hardware in the AI ecosystem.
Source: Firstpost · Jean Leon
Anthropic’s early-stage chip talks with Samsung signal a hardware pivot among AI startups. With a $61.5B valuation, the move highlights how venture-backed firms are building competitive moats through custom silicon.
Anthropic’s potential custom chip with Samsung signals a move toward differentiated cloud infrastructure that could alter the economics of AI for SaaS providers, as Amazon and Google already offer custom silicon.
Anthropic’s early Samsung chip talks aim to reduce Nvidia’s 80% market share dominance in AI accelerators, following OpenAI’s Jalapeño. The shift toward workload-specific silicon could unlock new model efficiencies.
Source: Fp Tech Desk (in) · Lucas Ropek (us)
Nvidia and Valar Atomics have demonstrated a nuclear microreactor powering Blackwell GPUs in a Utah data center, eliminating water-based cooling entirely. This breakthrough could solve the massive water consumption problem of AI infrastructure and set a new standard for sustainable AI deployments.
Source: Miamiherald · Kansascity
OXMIQ, the AI chip startup from Raja Koduri, has raised a $35M Series A round, taking total funding to $60M. The company plans to license a unified AI chip architecture, challenging Nvidia’s dominance and targeting sovereign AI programs, with a commercial launch expected by early 2027.
Source: Shilpa Ranipeta (in) · Max A. Cherney (my)
A sharp tech sell-off dragged the Nasdaq to a 5% weekly loss on June 26, led by a 10%+ plunge in Sandisk and a 7% drop in Micron. Inflation fears and Jeremy Grantham's market bubble warning fueled a rotation to gold and bonds.
Source: The Motley Fool · fool.com
AI semiconductor stocks suffered a sharp repricing on Friday after OpenAI’s IPO delay raised fears of softening chip demand. Broadcom, AMD, Micron, and Sandisk all dropped, snapping a multi-year rally that has propelled the PHLX Index up 429%. The episode highlights the acute sensitivity of high-beta AI plays to funding sentiment and suggests the AI trade may be entering a more discriminating phase.
The sharp sell-off in AI chip stocks after OpenAI’s IPO delay highlights the AI industry’s deep dependence on a handful of well-funded labs. With OpenAI pausing its public entry, the pace of infrastructure expansion for next-generation models could moderate, posing risks for both chipmakers and the AI research community. Yet the long-term demand story remains intact, even if the near-term funding pipeline looks choppier.
Source: CPA (us) · fool.com
OpenAI and Broadcom's custom inference chip, Jalapeño, aims to slash recurring AI costs amid soaring industry-wide capital expenditures and debt. Broadcom's AI revenue still projects $100B long-term, but the stock's recent 21% drop signals investor wariness over the AI buildout's price tag.
Source: Sacbee · Kansascity
Micron’s strong earnings guidance and Qualcomm’s 2029 data center target provide fresh proof that AI infrastructure spending is not a bubble but an accelerating megatrend. The $400 billion after-hours valuation jump reflects renewed confidence in AI's hardware foundation.
Qualcomm’s goal of $15 billion in data center sales by 2029 signals a massive expansion of chip supply for cloud AI workloads. For SaaS platforms, this means more accessible, powerful compute that could reshape infrastructure costs and product capabilities.
A staggering $400 billion in semiconductor market cap returned in after-hours trading after Micron’s blowout forecast and Qualcomm’s $15B data center target directly countered Tuesday’s AI sell-off. The rally highlights the sector’s extreme sensitivity to guidance and the enduring faith in AI infrastructure spending.
Source: Noel Randewich (my) · Channelnewsasia
OpenAI partnered with Broadcom to design a bespoke AI inference chip from scratch in just nine months, a timeline that defies industry norms. This speed showcases a new model of agile chip development, potentially inspiring AI startups to pursue custom silicon.
The new Jalapeño chip's 'substantially better' performance per watt promises to drive down inference costs, a critical factor for SaaS platforms embedding LLM capabilities. This could lead to more affordable AI features and new SaaS pricing models.