48 articles
Huawei's 950PR AI chip designed for Nvidia CUDA compatibility marks China's strategic response to US export restrictions on advanced semiconductors. Companies are restructuring supply chains and accelerating domestic production capabilities as geopolitical tensions force the industry to operate across fragmented technology ecosystems.
Arm Holdings is shifting from pure IP licensing to direct chip manufacturing, projecting $15 billion in annual revenue from AI data center processors within five years. The company's new Arm AGI CPU delivers over 2x performance per rack versus x86 platforms, positioning the semiconductor designer to capture infrastructure spending during industry consolidation.
ARM Holdings is launching a data center chip business targeting $15 billion in annual revenue within five years, marking a shift from pure IP licensing to silicon sales. The ARM AGI CPU delivers more than 2x performance per rack compared with x86 platforms as partners seek scalable ARM deployment for AI infrastructure.
NVIDIA's BioNeMo platform has secured strategic partnerships with pharmaceutical giants Eli Lilly and Thermo Fisher Scientific to power AI-driven drug discovery infrastructure. The platform is enabling multiple biotech AI startups including Terray, Apheris, Natera, and Basecamp Research to build foundation models, replicating NVIDIA's GPU dominance strategy in the high-value biotech market.
NVIDIA's BioNeMo platform is gaining adoption among major life sciences companies for AI-driven drug discovery. Thermo Fisher and Eli Lilly are among the early partners deploying the platform. AI-native biotech firms are building specialized foundation models for biological research using NVIDIA's infrastructure.
This article cannot be written without source documents. The anti-hallucination rules require all facts, dates, and citations to come from explicitly provided source documents.
Rivian no longer expects to achieve profitability by 2027 due to substantial spending on autonomous vehicle technology. The decision follows an Uber-Rivian robotaxi partnership announcement and Uber's investment in the EV manufacturer, highlighting the capital-intensive nature of autonomous vehicle development.
NVIDIA is positioning its BioNeMo platform as foundational infrastructure for AI-driven drug discovery through partnerships with pharmaceutical giants including Eli Lilly and Thermo Fisher. The company is establishing itself as critical middleware between AI compute and lab workflows, while enabling a new ecosystem of biotech AI startups through domain-specific tooling.
NVIDIA secured integrations with Salesforce, Adobe, Atlassian, and Siemens for its Agent Toolkit and OpenShell platform, positioning itself as the infrastructure layer for enterprise agentic AI. The move comes as competitor Yann LeCun raises over $1 billion and regulatory pressure targets Anthropic, marking a shift from research AI to production deployment.
Multiple vehicle manufacturers are preparing to launch Level 2++ autonomous driving systems as standard features across their lineups in 2025-2026. The deployment wave aims to expose consumers to advanced driver assistance at scale while generating real-world training data for higher autonomy levels.
NVIDIA unveiled three industry-specific AI platforms on March 16, targeting space computing, warehouse automation, and semiconductor design. The Space Computing Platform and Space-1 Vera Rubin Module mark the company's entry into orbital computing infrastructure. Partnerships with KION, Siemens, and Accenture extend AI capabilities into logistics operations.
The three major cloud hyperscalers are competing for enterprise AI dominance through platform expansions and strategic partnerships. Microsoft's Azure OpenAI Services, Google's Vertex AI, and AWS Bedrock now offer competing enterprise AI infrastructure stacks. Analyst upgrades of NVIDIA, Dell, ASML, and Microsoft signal institutional confidence in the AI buildout cycle.
The semiconductor industry is dividing into AI winners and traditional memory losers as companies position for divergent growth trajectories. STMicroelectronics, Nordic Semiconductor, and NXP are backing the new Aliro 1.0 standard for interoperable access control, while Lattice Semiconductor forecasts Q1 revenue of $158-172 million amid sector uncertainty. Wolfspeed and InspireSemi are doubling down on automotive silicon carbide and AI accelerators, respectively.
The casual steakhouse chain reopened its first location in 2024 after shutting all 280 restaurants in 2008. Consumer memory gaps and trust erosion create severe customer acquisition challenges for the 59-year-old brand founded by Norman Brinker.
KLA Corp. forecasts mid-to-high teens growth in advanced packaging equipment for calendar 2026, outpacing broader semiconductor industry growth. The packaging equipment maker expanded indium phosphide capacity by over 20% in Q4 alone as AI chip manufacturers race to improve interconnect density. Advanced packaging is emerging as a competitive moat for chipmakers unable to access leading-edge process nodes.
Nvidia allocated $2B across partnerships with Coherent and Lumentum for optical interconnect technology targeting AI data center workloads. Apple and Samsung launch competing silicon in March 2026—M5 Pro/Max chips for MacBook Pro and Galaxy S26 processors—as semiconductor demand accelerates across industrial and data center segments.
Meta's 200-language AI model and OpenAI's expansion have pushed investors to abandon African language startups, according to AI ethics researchers Timnit Gebru and Abeba Birhane. The researchers claim Big Tech uses 'AI for good' rhetoric to deflect criticism while threatening smaller competitors with obsolescence and low-ball data acquisition offers.
Nvidia is investing $4 billion in photonics integration to address AI computing's energy demands, while silicon carbide and GaN semiconductors gain traction in EV and datacenter markets. Emerging players like InspireSemi and Credo are deploying novel architectures that prioritize power efficiency over raw performance, marking a fundamental shift in semiconductor design priorities.
Google released its seventh-generation Ironwood TPU while Anthropic committed to 1 million Amazon Trainium2 chips, signaling a shift away from general-purpose GPUs. Hyperscalers are building custom AI accelerators to reduce inference costs as cloud AI workloads scale. Analysts predict custom chips could capture 25-30% of the AI training market by 2027.
Meta secured a large-scale GPU procurement contract with AMD on February 25, 2026, sending AMD shares up 10%. The deal marks a strategic shift in enterprise AI infrastructure purchasing, as hyperscalers adopt multi-vendor strategies to reduce dependency on single suppliers and improve negotiating leverage.