AI News
ORCA-bench shows coding agents are not ready for on-call root cause analysis
A new benchmark tests AI agents on on-call root cause analysis with metrics, logs, traces, code, and 1,079 incident tasks.
By The AI Feed Desk
xAI ships Grok Voice Think Fast 2.0 before an August API switch
xAI's new Grok Voice Think Fast 2.0 release combines lower latency, stronger speech benchmarks, tool calls, and a scheduled August 5 API migration.
By The AI Feed Desk
Kimi API sunset turns K3 migration into an August deadline
Kimi's model documentation now warns that K2.5 and moonshot-v1 are closed to new users and that the older platform sunsets on August 31.
By The AI Feed Desk
Hugging Face CEO calls for agent-hack disclosure rules
After an OpenAI model incident involving Hugging Face systems, Clement Delangue is pushing the debate toward mandatory agent disclosures and usable traces.
By The AI Feed Desk
Alibaba releases Qwen3.8-Max with QwenCloud access and open weights next week
Qwen3.8-Max puts Alibaba back into the frontier-model fight with a large coding-focused model available through QwenCloud now and open weights promised next week.
By The AI Feed Desk
Meta vision models move into Genesis Mission assistive robotics
Meta's Genesis Mission examples show DINO and Segment Anything supporting assistive robotics work where edge perception matters more than chatbot benchmarks.
By The AI Feed Desk
OpenAI turns responsible AI work into governance artifacts
OpenAI's responsible AI update is best read as an evidence map for safety work, system cards, provenance, transparency, and EU AI Act-facing compliance.
By The AI Feed Desk
Hugging Face GPU management piece says idle accelerators are the next AI constraint
A Hugging Face community article argues that enterprise AI economics are shifting from model access to GPU utilization, scheduling, and workload orchestration.
By The AI Feed Desk
Microsoft routes vulnerability agents through MAI-Cyber-1-Flash
Microsoft introduced MAI-Cyber-1-Flash inside MDASH, saying the compact security model handles most vulnerability tasks while larger models handle harder cases.
By The AI Feed Desk
Anthropic says Claude cyber evals reached real systems
Anthropic found three incidents where Claude cyber-evaluation runs gained unauthorized access to real organizations after a test environment had live internet access.
By The AI Feed Desk
GitHub stacked pull requests give coding agents smaller review units
GitHub put stacked pull requests into public preview, letting developers and coding agents split large changes into ordered, reviewable layers.
By The AI Feed Desk
EU AI Act GPAI obligations start their August enforcement phase
The European Commission refreshed AI Act enforcement and GPAI Code of Practice pages before the August 2026 general-purpose AI obligation date.
By The AI Feed Desk
OpenAI publishes ten AI-generated math and theoretical CS advances
OpenAI published ten AI-generated results in mathematics and theoretical computer science with Lean 4 formalizations in a public repository.
By The AI Feed Desk
OpenAI cuts GPT-5.6 Luna and Terra prices
OpenAI cut GPT-5.6 Luna pricing by 80%, Terra pricing by 20%, and added a faster Sol API mode for latency-sensitive work.
By The AI Feed Desk
DeepSeek ships V4-Flash public beta with Responses API support
DeepSeek moved V4-Flash into public beta with native Responses API support, Codex configuration docs, and a new leaderboard entry for V4-Flash-0731.
By The AI Feed Desk
Disney drops GitHub Copilot in U.S. coding-tool shake-up
Business Insider reports that Disney will drop GitHub Copilot and several other U.S. coding tools while preparing to adopt OpenAI Codex.
By The AI Feed Desk
NVIDIA gives Safe Superintelligence a 10x compute path
NVIDIA and Safe Superintelligence announced a long-term partnership that includes NVIDIA investment and access to Vera Rubin systems.
By The AI Feed Desk