AI
Sakana AI Releases RL Conductor to Orchestrate GPT-5, Claude
Sakana AI released RL Conductor, a 7B parameter model that automatically orchestrates GPT-5, Claude, and other LLMs, outperforming individual models while reducing…
More AI news — page 33
Older Stories page 33 of 141
Enterprise
TokenArena Benchmark Reveals 12.5-Point Accuracy Gaps
New TokenArena benchmark reveals AI model performance varies up to 12.5 points across different deployment endpoints, while workload-specific pricing analysis shows dramatic…
AI Agents
DeepSeek V4, Poolside Laguna, SenseTime U1 Lead Major AI Model Week
DeepSeek released V4 with extended context processing, Poolside launched Laguna XS.2 for agentic coding, and SenseTime unveiled U1 for direct image processing,…
OpenAI
Poolside Launches Laguna XS.2 as Open Source AI Models Gain Traction
Poolside launched its open source Laguna XS.2 coding model while SenseTime released its U1 image model, as image AI drives 6.5x more…
AI
xAI Grok 4.3 Brings Chain-of-Thought Reasoning at $1.25/M Tokens
xAI launched Grok 4.3 with integrated chain-of-thought reasoning at $1.25 per million tokens, while DeepSeek-V4 delivers near state-of-the-art reasoning performance as open-source.…
AI
AI Architecture Advances: Efficiency Breakthroughs Drive
AI architecture advances in 2026 focus on efficiency over scale, with Zyphra's 8B-parameter MoE model, Subquadratic's claimed 1,000x efficiency gains, and Google's…
OpenAI
AI Safety Research Tackles Sycophancy and Misalignment Risks
New AI safety research identifies sycophancy as a boundary failure between social alignment and epistemic integrity, while separate work explains emergent misalignment…
SGI
AGI Research Shows Patterns
Research reveals that advanced AI reasoning models are converging toward identical representations of reality, while new efficient models like ZAYA1-8B demonstrate competitive…
AI Agents
Microsoft Agent 365 Goes Live as Enterprise AI Agent
Microsoft moved Agent 365 to general availability amid rising enterprise AI agent security concerns, with 88% of organizations reporting agent-related incidents. Anthropic…
Companies
NVIDIA Invests $40B in AI Infrastructure Partnerships
NVIDIA has committed over $40 billion in strategic investments this year, including major partnerships with Corning and IREN to secure AI infrastructure…
Security
CVE-2026-41940 cPanel Flaw Exploited as Zero-Day for 30 Days
Critical vulnerabilities in cPanel, Linux kernel, Android, and AI development tools are under active exploitation, with the cPanel flaw being exploited as…
OpenAI
AI Reasoning Models Drive Up Compute Costs
AI reasoning models like OpenAI's o1 and xAI's Grok 4.3 deliver improved performance but create significant cost challenges through hidden reasoning tokens…
