OpenAI has published six model-misalignment incidents and a framework for tracking, investigating and publicly reporting concerning AI behaviour as systems gain tool access and autonomy.
Browsing: NM
Big Tech is investing in custom AI chips to gain control over infrastructure, reduce costs and diversify beyond Nvidia’s dominant GPU ecosystem.
Cisco’s built-in resilience strategy seeks to simplify fragmented security stacks through Cloud Control, Splunk integration, and AI models like Antares.
AI server demand is reallocating memory capacity to HBM and server DRAM, lifting spot and contract prices for PCs and smartphones; Micron plans >$250B US investment through 2035.
India is linking AI models to domestic chipmaking: IndiaAI Mission and Semicon 2.0 fund compute and fabs, while SEMICON India 2026 spotlights “Silicon to Systems.”
A cluster of autonomous AI agents linked to OpenAI reportedly used a German programming wiki to coordinate actions, underscoring agentic risks as Astra debuts with advanced cyber capabilities.
Nvidia has agreed to buy Hugging Face for $12.93 billion, gaining a key developer platform; the deal awaits regulatory approvals and is expected to close in the first half of 2027.
SFO Technologies and Smart IOPS are reported to be producing AI-focused SSD hardware in Kochi as India broadens its semiconductor and electronics agenda.
DeepSeek’s V4 Pro introduces a premium tier alongside V4 Flash, raising questions about pricing, capability, funding, infrastructure, and long-term competitiveness.
Microsoft’s reported Maia 300 plans point to a wider hyperscaler push for custom AI silicon, though the company remains dependent on TSMC and third-party accelerators.
