Every story we have kept, newest first.
Looking for the daily editions? → Past editions
Page 9 of 1372
The benchmark compares Qwen3-Coder-30B and NVIDIA Nemotron-3-Nano-30B across G5, G6, G6e, and G7 instances on Amazon SageMaker AI. It measures throughput, latency, and cost per token, highlighting price-performance gains from G7 NVIDIA Blackwell GPUs.
AWS Machine Learning Blog · 9d ago
Google Cloud is expanding its enterprise AI strategy through a deal with Accenture. The partnership will use forward-deployed engineers to help customers adopt AI and overcome implementation bottlenecks.
TechCrunch AI · 9d ago
HPE Zerto built an agentic troubleshooting system powered by Amazon Bedrock that operates on-premises within customer environments. The system uses a multi-agent architecture and Strands Agents to ground troubleshooting in live disaster recovery data.
AWS Machine Learning Blog · 9d ago
China's official average daily token consumption reached 500 quadrillion by the end of June, but the article warns that comparisons with the US remain uncertain because providers use incompatible scopes and may double-count or omit workloads. Token volume also does not directly translate into AI revenue, as model quality, latency, workload value, and infrastructure economics matter.
虎嗅 · 9d ago
DiDi built a transparent, self-owned contact center quality assurance system on Amazon Bedrock to replace an opaque third-party tool. Intent verification accuracy increased from 38% to 86%, compliance scoring exceeded 90%, and Voice of Customer analysis became much faster across Spanish and Portuguese support.
AWS Machine Learning Blog · 9d ago
The UK government plans to require companies including Apple and Google to prevent children from taking, sharing, or viewing nude images on phones and tablets. Culture Secretary Lisa Nandy announced the policy in the House of Commons.
Bloomberg Technology · 9d ago
China has imposed a cash deposit of up to 99.2% on imports of a Japanese chemical used by chip fabs to create films inside logic and memory chips. The measure took effect Tuesday after China’s Ministry of Commerce issued a preliminary determination.
The Next Web (TNW) · 9d ago
WhatsApp is reportedly developing native support for third-party AI agents, allowing users to connect them to private chats through an API key. Agents will be isolated from group chats, contacts, files, and other conversations, while WhatsApp will warn that their responses may be inaccurate.
IT之家 · 9d ago
Google and Accenture are reportedly placing 1,000 engineers inside client offices to help organizations turn AI commitments into working implementations. Accenture CEO Julie Sweet said clients increasingly need hands-on assistance rather than another AI sales pitch.
The Next Web (TNW) · 9d ago
SoftBank Group will meet investors in New York next week to gauge interest in a potential US dollar junk-bond sale. The borrowing would help finance Masayoshi Son’s aggressive investment push in artificial intelligence.
Bloomberg Technology · 9d ago
Google is accelerating Chrome’s release schedule to deliver security patches and new features every two weeks. The change reflects a security landscape increasingly shaped by AI.
TechCrunch AI · 9d ago
TSMC plans to begin using High-NA EUV lithography in 2030, with its A10 or A11 process technologies considered likely candidates. The company also expects new scanners supporting 6×12-inch photomasks by 2033.
Tom's Hardware · 9d ago
Franklin Templeton strategist Katrina Dudley says returns from AI spending may take longer to materialize. She argues that future infrastructure investment will depend on clearer evidence of AI’s financial returns.
Bloomberg Technology · 9d ago
Huawei is reportedly coordinating China’s effort to develop a domestic alternative to ASML’s chipmaking machines. However, the flagship system still relies on German-made Zeiss lenses, according to a Financial Times report cited by TNW.
The Next Web (TNW) · 9d ago
The Hugging Face article examines a more precise approach to AI safety: refusing only the harmful or disallowed portion of a topic rather than rejecting the entire subject. This approach could make models more useful while preserving appropriate safety boundaries.
Hugging Face Blog · 9d ago
DeepMind’s AlphaGenome Atlas provides a predictive map of the molecular effects of 9 billion possible single-letter DNA variants across the human genome. The resource could help researchers investigate how genetic changes influence biological function.
DeepMind Blog · 9d ago
Runway AI’s annual recurring revenue reached $200 million in September. The milestone reflects growing business demand for its image- and video-generation software used in marketing materials and advertisements.
Bloomberg Technology · 9d ago
Celero Communications has raised $275 million in a funding round valuing the networking-chip startup at more than $3 billion. Alphabet’s CapitalG backed the round as demand grows for more efficient AI data centers.
Bloomberg Technology · 9d ago
Claude can use Gmail capabilities to help send emails on a user's behalf. Users who are concerned about mistakes can configure Claude to request approval before taking action.
Engadget · 9d ago
Google DeepMind unveiled AlphaGenome Atlas, an AI platform designed to predict the effects of every possible single-letter change in the human genome. The researchers say the tool could accelerate biological research and help identify paths toward new disease treatments.
The Verge · 9d ago
Microsoft is reportedly preparing another record-setting Patch Tuesday as AI models rapidly discover vulnerabilities in widely used software. The trend follows Anthropic’s Mythos model finding flaws across major operating systems and browsers, followed by OpenAI’s cybersecurity-focused model.
The Verge · 9d ago
IQE CEO Jutta Meier warned that China controls 70% of a material needed by the AI industry, creating a supply-chain vulnerability. The Cardiff-based Welsh chipmaker recently reported a 40% revenue increase, while its shares have risen 849% this year.
The Next Web (TNW) · 9d ago
Ifanr reports that Vivago R1 has launched globally and tests its ability to create an AI-generated short film. The hands-on project uses a cyberpunk interpretation of Journey to the West to demonstrate accessible AI video creation.
Ifanr (爱范儿) · 9d ago
Tom's Hardware benchmarked the newly released Qwen 3.8 27B across an RTX 5090 and other hardware. The testing indicates that sufficient VRAM alone cannot compensate for serious software and inference-engine bottlenecks.
Tom's Hardware · 9d ago
OpenAI has released a new AI agent that reportedly behaved unpredictably soon after launch. The excerpt provides no model name or technical specifications, but the incident raises concerns about agent reliability and safeguards.
The Guardian Technology · 9d ago
China’s August exports rose 25.0% year over year, while imports increased 28.2%, with AI infrastructure demand, semiconductor prices, and new-energy vehicle exports providing major support. Chip exports jumped 129.8%, but analysts expect growth to moderate as trade barriers, higher base effects, and the semiconductor cycle create headwinds.
虎嗅 · 9d ago
Qualcomm has signed Amazon as a data center chip customer for a deal spanning multiple generations. Amazon will also have the right to acquire up to $4 billion in Qualcomm shares.
Bloomberg Technology · 9d ago
Cloudflare's Automatic Key Exchange probes TLS 1.3-capable customer origins to identify supported key agreement algorithms. It then prioritizes the most secure option, including post-quantum connections when available, across 45 billion daily connections.
Cloudflare Blog · 9d ago
Ben's Bites presents a report about the first GPT-6 model and describes what the author is building with it. The excerpt provides no benchmark, access, pricing, or technical implementation details.
Ben's Bites · 9d ago
Arm has launched Neoverse CSS N4, a server CPU platform designed to address rising compute demand in the agentic AI era. Each die can support up to 128 cores.
InfoQ中国 · 9d ago