From Hacks to Bioweapons, Claude Misuse Is Now In every single place

In a single case, a bunch of Russian state-sponsored hackers recognized by Microsoft as Midnight Blizzard used Claude for reconnaissance, breaching targets that included Ukrainian and different European authorities networks, and stole information and maintained entry. Cybercriminal group ShinyHunters used Claude in virtually each stage of its hacking and extortion campaigns. Disinformation campaigns specializing in…

Read Full News

Can LLMs Engineer Their Own Agent Harness? ByteDance Seed’s HarnessDev Says Only 34 of 64 Changes Generalize

An agent harness is the code around a model: execution loop, tools, context, state, recovery, and verification. Per the Terminal-Bench 2.1 leaderboard, GPT-5 solves 35.2% of tasks inside Terminus 2 but 49.6% inside Codex CLI with identical weights. Most benchmarks keep that harness fixed. HarnessDev proposed by team of researchers from ByteDance Seed, Singapore University…

Read Full News

Y Combinator’s Garry Tan desires US open-weight AI labs to ‘distill’ frontier fashions, too

In relation to Chinese language AI labs utilizing distillation methods to extract information from frontier mannequin makers, Y Combinator CEO Garry Tan is hoping regulators keep out of it. In actual fact, he thinks U.S. AI labs ought to maybe play the identical recreation. “I might do nothing,” he informed CNBC in an interview earlier…

Read Full News

Khosla Ventures is opening a New York office this fall — its first outpost outside Sand Hill Road

Keith Rabois has spent most of his 13-year career in venture capital associated with Khosla Ventures’ longtime home in Menlo Park, California. But that’s changing. Speaking Thursday night at TechCrunch’s StrictlyVC event in New York’s West Village, Rabois confirmed that Khosla Ventures is opening its first-ever office outside Sand Hill Road. The new outpost will…

Read Full News

Anthropic Adds Plugin Evals to Claude Code: 6 Grader Types, a No-Plugin Baseline, and a CI Gate for Skills

Anthropic has published a new plugin evals workflow for Claude Code. The claude plugin eval command runs a plugin against realistic prompts, grades what Claude produced, and compares the result with a run where the plugin is not loaded. It answers 3 questions plugin developers could not previously measure: does the skill trigger, does it…

Read Full News