OpenAI Restarts Frontier Training After Its Own Agents Spent Billions of Tokens Testing the Locks
OpenAI says teams of its own AI agents spent hundreds of billions of tokens trying to escape its training sandboxes before it restarted frontier training. Along the way, a 9.8-rated web bug surfaced.
The post OpenAI Restarts Frontier Training After Its Own Agents Spent Billions of Tokens Testing the...
The focus on AI safety disasters ignores everyday harms
Consider what passes for success in technology policy. An AI company promises that its chatbot will not encourage suicide. A social media platform makes it more difficult for children to buy fentanyl. These are important protections, and they should have existed all along. But companies can clear th...
Tsinghua AI Open-Sources AKV, a Cache Engineering Server for LLM Agents on SGLang
Tsinghua researchers released AKV, an SGLang-based serving system for cache engineering. LLM agents drop and reposition KV cache instead of summarizing history, cutting average cost 27.7% across 4 agent benchmarks.
The post Tsinghua AI Open-Sources AKV, a Cache Engineering Server for LLM Agents on S...
Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model
Nanbeige LLM Lab released Nanbeige4.3-33B, a 33B dense Apache 2.0 model with 512K context. It reports 80.9 on Terminal-Bench 2.1 and will teach the upcoming Nanbeige4.3-3B.
The post Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model appeared first on MarkTechPost.
Hugging Face Releases Carbon-X, a 14B Open DNA Language Model that Brings DNA and Natural Language Together for the First Time
Hugging Face released Carbon-X, an open 14.8B DNA language model built on Qwen3-14B. It reads raw DNA in chat, answers in English, and averages 66.7% across 5 DNA benchmarks under Apache 2.0.
The post Hugging Face Releases Carbon-X, a 14B Open DNA Language Model that Brings DNA and Natural Language ...
Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster
Stanford and UC Berkeley researchers released TokenGate, a drop-in learned token pruning module for LLMs. It cuts long-context prefill time-to-first-token by 1.4x to 11.3x and peak memory by 1.5x to 3.4x without accuracy loss.
The post Researchers from Stanford and UC Berkeley Introduce TokenGate: T...
Anthropic Launches OSS Scanner, a Free AI Vulnerability Scanner
Anthropic has launched OSS Scanner, a free, opt-in vulnerability scanner for critical open-source projects. The Anthropic OSS Scanner runs the company’s strongest models, including Claude Mythos, against enrolled codebases. Reports go straight to maintainers without human review. That trade delivers...
Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,024 Agents
Microsoft Research open-sourced Agensh, a multi-agent harness with no central orchestrator. Self-organizing workers share a Git repo, chat and a findings board, scaling from 1 to 1,024 agents on ProgramBench under a 6-hour budget.
The post Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,0...
Cognition Open-Sources an Agent Plugin that Helps Secure Systems against Recent Cybersecurity Threats
Cognition released the Devin Vulnerability Fixes plugin, 3 MIT-licensed agent skills that inventory, patch and test PHP, Node.js and self-managed GitLab deployments hit by this week’s critical RCE flaws, then open 1 PR per repo.
The post Cognition Open-Sources an Agent Plugin that Helps Secure Syste...
Forced to pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.
A new study found Grok's AI overwhelmingly recommends Republican candidates while OpenAI's GPT favors Democrats, with the two AI models disagreeing in more than 90% of nearly 500 election races.
Meet Anthropic’s sandbox-runtime: OS-Level Sandboxing for AI Agents, No Containers
Anthropic sandbox-runtime (srt) is the open-source, OS-level sandbox behind Claude Code. It restricts file and network access for agents, MCP servers and CLI tools with Seatbelt on macOS and bubblewrap on Linux, no containers needed.
The post Meet Anthropic’s sandbox-runtime: OS-Level Sandboxing for...
Stateful Agentic Engineering Copilots with Letta: Core Memory, Shared State, Semantic Retrieval, Custom Tools, Streaming, and Portable Agent Workflows
Discover how to extend agentic memory beyond basic prompts by building a stateful engineering copilot with Letta. This comprehensive guide covers inspectable core memory, archival vector retrieval, multi-agent shared state, custom tools, and portable workflows.
The post Stateful Agentic Engineering ...
NVIDIA Releases Nemotron 3.5 Bolt E4B 4GB: An Open Small Language Model for Local Voice Agents, Game Characters and Robots
NVIDIA released Nemotron 3.5 Bolt E4B 4GB, an open small language model post-trained from Gemma 4 E4B for local voice assistants, game characters and robots, targeting about 4GB of VRAM on RTX and Jetson.
The post NVIDIA Releases Nemotron 3.5 Bolt E4B 4GB: An Open Small Language Model for Local Voic...
Kyutai CADRE: Video Autoencoders That Cut World Model Lag Below 230 ms
Kyutai released CADRE, a family of 7 open causal diffusion video autoencoders for interactive world models. World models built on CADRE run at 20 fps on 1 H100 and react in under 230 ms.
The post Kyutai CADRE: Video Autoencoders That Cut World Model Lag Below 230 ms appeared first on MarkTechPost.
Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model with 1M Context
Moonshot AI released Kimi K3.1 Pro, a 2.8T-parameter, 104B-active MoE model with 1M-token context. It scores 65.7% on Terminal-Bench 4.0, costs $2/$10 per million tokens, and open weights arrive by Nov 4, 2026.
The post Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model ...
From Toast to Bessemer: Who founders will hear from at TechCrunch Founder Summit 2026
Hosted by TechCrunch in partnership with Fidelity Private Shares, the full-day event goes beyond fundraising and AI to explore the realities of finding customers, building teams, sharpening a pitch, developing founder leadership, and scaling companies both inside and outside traditional startup hubs...
Hugging Face’s Carbon-X Lets Researchers Chat With DNA
Hugging Face's open Carbon-X model reads raw DNA and answers in plain English, nearly matching frontier models on DNA benchmarks at a fraction of the cost.
The post Hugging Face’s Carbon-X Lets Researchers Chat With DNA appeared first on DailyAI.
TinyFish Breaks Down Agentic Search Architecture for Production AI Agents
TinyFish published two engineering research articles on agentic search. One defines a 7-phase production architecture with evidence contracts and bounded retries. The other explains why its deep research system dropped a separate planner and per-query subagents.
The post TinyFish Breaks Down Agentic...
Google Cloud Launches Agent Compute Engine: A MicroVM per AI Agent
Google Cloud announced Agent Compute Engine, a compute service that gives every AI agent its own hardware-isolated microVM with no credentials inside. It already runs Gemini agents and enters preview in the coming months.
The post Google Cloud Launches Agent Compute Engine: A MicroVM per AI Agent ap...
You.com launched Inference Router, an OpenAI-compatible API for 12 open-weight models with a 99.9% uptime SLA, zero data retention, 100% US hosting, server-side web search and no platform fee.
The post You.com Launches Inference Router: 12 Open-Weight Models With 99.9% Uptime SLA appeared first on M...