Sam's News — ai — 2026-09-08¶

Security¶

7.5 Google warns attackers built full cyberattack campaign using AI in six hours¶

Google reported that threat actors constructed a complete cyberattack campaign in six hours leveraging artificial intelligence tools.

Sources: Techzine Global RSS

AI Policy¶

7 OpenAI chief scientist advocates for slowing AI research pace¶

OpenAI's chief scientist called for a slowdown in artificial intelligence research, raising concerns about development speed and safety.

Sources: SiliconANGLE RSS, Fortune RSS

Hardware¶

7 AMD Ryzen AI Max PRO 400 runs 300 billion parameter LLMs with 192GB memory¶

AMD's Ryzen AI Max PRO 400 processor can run 300-billion-parameter large language models locally with 192GB integrated memory.

Sources: shattered.io RSS Update to: AMD Ryzen AI Max PRO 400 Targets 300B LLM Inference With 192GB Memory

6 SEMIFIVE Begins Mass Production of HyperAccel's Bertha AI Inference Accelerator¶

SEMIFIVE has commenced mass production of HyperAccel's Bertha AI inference accelerator chip on Samsung's 4nm process node.

Sources: PR Newswire RSS

Policy¶

6.5 Trump administration divided over AI training rights on copyrighted American work¶

The Trump administration is internally split over whether AI companies should be permitted to train models on Americans' copyrighted works.

Sources: Heartlander News RSS

AI¶

6.5 DeepSeek Plans Major Engineering Expansion as AI Competition Shifts to Systems¶

Chinese AI company DeepSeek announces hiring of 150 engineers as the competitive focus moves from model development to production systems and applications.

Sources: Global Times RSS

6 Tencent Releases Open-Source Hy4 LLM With 770B Parameters¶

Tencent has released a preview of Hy4, an open-source large language model with 770 billion parameters and 1 million token context window.

Sources: Gadget Pilipinas RSS

5.5 AI models' neurological impact on human cognition examined¶

Research explores how AI systems are affecting human brain function and cognitive patterns.

Sources: MediaPost RSS

5.5 Hyundai Card deploys generative AI for PR and executive training¶

Hyundai Card launched generative AI systems for public relations and large language model training for executives.

Sources: FF News RSS

5.5 IBM Research llm-d optimizes existing hardware for large language model inference¶

IBM introduced llm-d, a tool that maximizes LLM performance on existing enterprise hardware without requiring upgrades.

Sources: IBM Research RSS