Cisco Antares Models Bring Local AI to Vulnerability Triage

Cisco’s Antares AI models localize potentially vulnerable code files on premises, but benchmark limits mean security teams still need human review.
SecurityWeek

Nuclear-Sabotage Malware Benchmark Trips Up Most Frontier AI Models

SentinelOne built a long-horizon reverse-engineering benchmark for frontier AI models using its own investigation into the Fast16 malware.
Tech Times

New AI Benchmark Holds GPT-5.5 at 43% on Cross-Domain Reasoning Chains

Relay-Bench, a new AI benchmark posted to arXiv in July 2026, chains problems across seven reasoning domains in a single ...

Google expands Gemini with cheaper models and a bug-hunter it keeps on a leash

Google LLC today launched three new Gemini Flash models and moved its CodeMender code-security agent into preview, part of a ...

China’s Moonshot AI releases Kimi K3, the largest open-source model ever, rivaling top U.S. systems

Moonshot AI released Kimi K3, a 2.8 trillion-parameter open-source AI model from China that rivals OpenAI, Anthropic and other top U.S. systems in frontier AI benchmarks.
The Heartland InstituteOpinion

Climate Change Weekly # 584—They Knew: Impossible Climate Model Scenario Widely Used Despite Known Flaws

The impossible RCP8.5 has been used as a reference case in more than 70 peer-reviewed articles despite its official ...
Ars Technica

Google updates Android Bench with new LLMs, but Gemini still lags behind

Code generation is emerging as one of the most popular applications for large language models (LLMs), but not all agents are equally good at all development tasks. Google created a benchmark earlier ...
ndtvprofit

Nasscom Bats For Testing India's Critical Systems With Open-Source AI Models Amid Uncertain Frontier Access

India should continue engaging with the US government for access to the world's most advanced artificial intelligence models, but at the same time aggressively test its own critical infrastructure ...
techtimes

Most AI Models Would Run Your Company Into the Ground, Princeton’s CEO-Bench Finds

Jian Feng Zhen (L), 23, and Dayi Deng, 23, (R), both from China, walk to class on the campus of Princeton University April 23, 2002 in Princeton, NJ. Twenty-six percent of the student population at ...
Forbes

Buckle Up: The Bad Guys Now Have A Model As Powerful As Mythos

This voice experience is generated by AI. Learn more. This voice experience is generated by AI. Learn more. China's Z.ai has released GLM-5.2, an open-weight AI model capable of repository-scale ...
CNN

US government allows Anthropic limited release of AI model that sparked cybersecurity concerns

The US government has allowed Anthropic to release its powerful Mythos AI model to select companies and organizations, revising license requirements after ordering an export block earlier this month ...
IEEE

Real-Time Detection of Mixed-Critical Events Using Vision-Language Models

Abstract: This paper addresses the detection and classification of mixed-critical events, such as fire incidents, traffic accidents, and violence-related events in urban surveillance systems. It ...