Skip to content

Scan record

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

Yesterday · Nov 1940 loaded

  1. Ooredoo Group concludes Digital Ecosystem Conference, forming draft Doha Principles to advance a trusted digital economy

    NewsOoredoo GroupGovernanceThe item you came from

  2. The Australian Securities and Investments Commission (ASIC) makes AI harm to consumers a 2027 enforcement priority

    NewsMLexGovernance

  3. Google says misconfigured test server exposed autonomous AI attack operation; system contained Russian-language comments

    NewsEscudo DigitalEvent

  4. Google AI agent, after receiving a "never give up" instruction, attempted hacking and impersonating employees

    NewsThe ElecEvaluation

  5. Jensen Huang and Nvidia are not a honeytrap story but a national security story

    NewsFortuneGovernance

  6. On the eve of Anthropic's listing, Sam Altman reiterates OpenAI's listing plan, having previously paused it over safety considerations

    News澎湃新闻 · 检索「AI安全」AI updates

  7. Microsoft agent-framework PyPI package poisoned

    NewsStepSecurityEvent

  8. When figures and text disagree: evaluating autonomous visual evidence verification in scientific papers

    PaperarXiv

  9. Semantics-aware contrastive and explicitly disentangled representations for AI-generated video detection

    PaperarXiv

  10. Log analysis for LLM agents: a lifecycle-oriented survey

    PaperarXiv

  11. IIJ Global and VeriCerts launch advertising trust gateway PoC: verifying ad issuers and tampering via DID/VC

    NewsIIJ Global SolutionsAI updates

  12. LLM jailbreak detection based on contrastive prototype structure

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  13. George Monbiot commentary: The UK growth plan forces through data centers nobody wants, betting that the AI bubble won't burst

    NewsThe Guardian · AIGovernance

  14. Researchers propose inducing monitorable reward hacking to compensate for the failure of gradient routing

    NewsLessWrongAlignment

  15. A LessWrong author posted about protesting Anthropic's IPO on November 19, calling for a ban on recursive self-improvement

    NewsLessWrongGovernance

  16. Shanghai AI Lab paper: switch to a different agent execution framework, and model score rankings may reverse

    NewsWeChat · Weijiqi LixinAI updates

  17. Tencent Zhuque Lab open-sources the pikit toolbox to systematically evaluate AI agent indirect prompt injection risk

    NewsWeChat · Weijiqi LixinTools

  18. Beihang and Alibaba teams propose the PASP framework, using multi-perception views and adversarial self-play to improve large models' ability to recognize disguised harmful content

    NewsWeChat · Weijiqi LixinDefenses

  19. Paper: under ANCHOR multi-turn attacks, refusal rates of 8 CLI agents drop to 0%

    NewsWeChat · Weijiqi LixinAttacks

  20. Xint discloses 5 Nginx RCE vulnerabilities, four discovered by automation

    NewsWeChat · SecureNexusLabEvent

  21. OpenAI is testing giving its autonomous agent Dots its own email address

    NewsiThome (Taiwan)AI updates

  22. Solomonoff induction based on reflective oracles

    NewsAI Alignment ForumAlignment

  23. CCN updates ENS information security policy and responsibilities and functions guidance, adding an appendix linking data protection and AI rules

    NewsCentro Criptológico NacionalGovernance

  24. Can you make a model believe false facts in the prompt?

    PaperarXiv: Alignment, Deception & Oversight

  25. Dependency fabrication in repository deployment agents

    PaperarXiv: Alignment, Deception & Oversight

  26. Nudge Security report: AI agents inherit employee access permissions, and their use is unsupervised

    NewsHelp Net Security AIEvent

  27. Negotiating control and accountability among stakeholders in high-risk AI application scenarios: development and evaluation of the ALIGN method

    PaperarXiv: 其他 AI 安全论文

  28. RECLAIM: robot vision-guided adversarial patch defense via embodied attention feedback

    PaperarXiv: 其他 AI 安全论文

  29. Truthgate: Checking coding agents' behavior against records they cannot edit—cheating, deterrence and disclosure on impossible tasks

    PaperarXiv: 其他 AI 安全论文

  30. PGPO: prototype-guided progressive obfuscation for privacy-preserving LLM-enhanced recommendation

    PaperarXiv: 其他 AI 安全论文

  31. Kaer: safety-aware replanning for multi-agent systems

    PaperarXiv: 其他 AI 安全论文

  32. Exactly stealing other clients' training data in federated learning

    PaperarXiv: 其他 AI 安全论文

  33. The limitation is in the input, not the model: state-dependent command safety in natural language interfaces for industrial controllers

    PaperarXiv: 其他 AI 安全论文

  34. Gradient-based defensive poisoning against system identification attacks on cyber-physical systems

    PaperarXiv: 其他 AI 安全论文

  35. GateWatch: Low-overhead LLM monitoring with Boolean circuits

    PaperarXiv: 其他 AI 安全论文

  36. SCR: Detecting Backdoor-Poisoned Samples via Clean Reference Residuals

    PaperarXiv: 其他 AI 安全论文

  37. Probing agent over-proactiveness in real-world settings with Continuation Audits

    PaperarXiv: 其他 AI 安全论文

  38. Nothing to Declare: Steering a Single Neuron Increases Concealment in Agent Teams

    PaperarXiv: 其他 AI 安全论文

  39. PIPETTE: prompt injection purple teaming for robust LLM agents

    PaperarXiv: 其他 AI 安全论文

  40. Can Weight-Space Programming Change a Model's Learned Ethical Priorities?

    PaperarXiv: 其他 AI 安全论文