Skip to content

Scan record

Everything the site collected, newest first by collection time (Beijing time), back seven days.

Nov 1840 loaded

  1. Chain-of-thought controllability reduces chain-of-thought monitorability

    PaperarXiv: 其他 AI 安全论文The item you came from

  2. Disentangling Moral Circuits and Moral Features in Language Models: Moral Content, Normative Judgment, or Cue Word? Evidence from 14 Models

    PaperarXiv: 其他 AI 安全论文

  3. Depth-Aware Activation Steering Based on Attention Residuals

    PaperarXiv: 其他 AI 安全论文

  4. Evaluating semantic insensitivity in financial sentiment analysis: implications for sustainable financial decision-making

    PaperarXiv: 其他 AI 安全论文

  5. From preference data to personalization: tracing sycophancy in LLMs

    PaperarXiv: 其他 AI 安全论文

  6. The defense rests: a court-inspired multi-agent LLM framework for automated safety argument generation

    PaperarXiv: 其他 AI 安全论文

  7. "Who will take over, and where the company goes from here": analysts expect the uncertainty facing investors to persist after Nvidia's earnings beat

    NewsCNBC · TechnologyAI updates

  8. Fireworks updates its investigation of two October 2026 security incidents; a small number of customer keys, models and datasets were accessed

    NewsFireworksEvent

  9. DarkRange says n8n MCP OAuth has a zero-click open redirect vulnerability

    NewsDarkRangeEvent

  10. NewCore says two Grok Bots made 31 uninstructed requests for enterprise data over five days, all blocked

    NewsNewCoreEvent

  11. OpenAI launches dual GPT-6.2 models, Astra first to reach the high-risk threshold for AI self-improvement

    NewsiThome (Taiwan)Evaluation

  12. After Jensen Huang's departure, interim CEO Cress shows NVIDIA remains at the center of AI

    NewsCNBC · TechnologyAI updates

  13. Microsoft and more than 40 organizations launch the AI Transparency Initiative to develop open standards

    NewsSource EMEAGovernance

  14. U.S. Secretary Malin said the recent system failures were not cyberattacks but stemmed from aging facilities and upgrade pressure

    News澎湃新闻 · 检索「AI风险」AI updates

  15. Jensen Huang is alleged to have had an extramarital affair with a Chinese spy

    NewsPlatformerAI updates

  16. Brazil's National Data Protection Authority (ANPD) AI regulatory sandbox receives an international award in Brussels

    NewsAgência Nacional de Proteção de DadosGovernance

  17. Coinbase says its AI intercepted $4.7 million in transfers that would have gone to a deepfake scam

    NewsDiarioBitcoinEvent

  18. Former OpenAI board member Helen Toner warns of AI self-improvement risk, saying the alignment challenge remains far from solved

    NewsTVNOpinion

  19. South Korea tightens exam security after AI cheating scandal

    NewsK-SélectionEvent

  20. Study shows some AI models change their answers when told they are being monitored

    NewsScience et vieAlignment

  21. California attorney general sues OpenAI over agent breach of Hugging Face

    NewsGIGAZINEGovernance

  22. Keynote speech by Singapore Senior Minister of State Tan Kiat How at the StableX Agentic Network launch event

    NewsMDDI (Singapore)Governance

  23. Gemini conversation retention period to be extended from 72 hours to 2 months

    NewsiThome (Taiwan)AI updates

  24. CAMH surveys 1,125 Canadians on AI literacy, releases Smarter Privacy Design Kit

    NewsOffice of the Privacy Commissioner of CanadaGovernance

  25. The Coruna and DarkSword iOS exploit kits in the Optimizely attack incident

    NewsStraikerEvent

  26. OpenAI board member Paul Christiano's personal position: frontier labs should pause some algorithmic progress

    XX @j_asminewangOpinion

  27. Theo published a first-hand review of Astra 6.2, criticizing the model

    XX @MinLiBuildsAI updates

  28. Team launches Horizon Labs: an automated AI R&D company run entirely by Claude

    XX @AmyPrbAI updates

  29. Orange Business adds a protection mode to Live Intelligence, letting enterprises choose their own AI environment

    NewsOrange BusinessAI updates

  30. IBM report: share of enterprise security incidents related to shadow AI rises from 20% to 43%

    NewsAD HOC NEWSEvent

  31. South Korean police: over the past year, 44.6% of online sexual violence suspects were adolescents

    Newssisats.comEvent

  32. As export controls tighten, NVIDIA announces it will stop selling H200 chips in China

    NewsUnite.AIAI updates

  33. OpenAI sued by California over safety evaluation agent "autonomously" breaching Hugging Face

    NewsSan Jose InsideGovernance

  34. People familiar with the matter say some Nvidia directors are pushing for Jensen Huang's return

    NewsSemaforAI updates

  35. Nvidia's startup holdings swell to a valuation of $65.5 billion, but Jensen Huang's departure casts doubt on future investments

    NewsTechCrunch AIAI updates

  36. Microsoft agent-framework compromised on PyPI: 1.26.1 and 1.26.2 carry credential-stealing code

    NewsChainguardEvent

  37. Trial of Alfozan, defendant in OpenAI eavesdropping case, set for January

    NewsCNBC · TechnologyGovernance

  38. CommBank launches CodeCheck to help Australian businesses find software vulnerabilities

    NewsCommonwealth Bank of AustraliaTools

  39. Europe's AI industry is not "all gloom," but it still lags behind

    NewsSemaforAI updates

  40. Correlation between behavior and reward is not sufficient for RL to reinforce that behavior

    NewsLessWrongAlignment