Skip to content

Scan record

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

Nov 1738 loaded

  1. OpenAI sued by California over hacking attacks launched by a runaway agent

    NewsThe TelegraphGovernanceThe item you came from

  2. Leading Chinese AI companies restrict cyberattack capabilities in new models

    NewsThe StarGovernance

  3. Malaysia prepares security and enforcement framework for AI governance bill, ensuring enforcement is possible on day one

    NewsThe StarGovernance

  4. Gemini 4 safety report: Carbon did not undergo a separate control evaluation; Argon bypassed safety monitoring up to 64% of the time when instructed to try

    NewsTechTimesEvaluation

  5. Redwood Research says Google reused Gemini 4 Argon's safety tests for Carbon

    NewsStartup FortuneEvaluation

  6. Asian and European stocks fall as tech slump fears intensify

    NewsSemaforAI updates

  7. Nvidia earnings overshadowed by CEO succession problem

    NewsSemaforAI updates

  8. Before the meeting with Xi Jinping, Trump's team splits over China policy

    NewsSemaforGovernance

  9. CAPES holds the sixth SIC Week, discussing AI risks and data protection

    NewsCAPESAI updates

  10. Anthropic IPO: How to subscribe, is it worth buying?

    NewsThe Guardian · AIAI updates

  11. Irish regulator halts X's training of AI on data collected by agents

    NewsPOLITICOGovernance

  12. Frag update: supports single-shot blind exfiltration and Base64 tokens

    XX @garethheyesAttacks

  13. California Attorney General sues OpenAI over misleading promotion; the complaint alleges it provided unsafe AI models

    NewsGolem.deGovernance

  14. Marius Hobbhahn clarifies: AI timelines have basically not changed in recent years; he only believes risks are more likely to appear earlier

    XX @MariusHobbhahnOpinion

  15. Belfield questions Christiano's AI takeover scenario

    XX @HaydnBelfieldOpinion

  16. OpenAI signs contracts with four independent evaluation organizations to conduct on-site safety evaluations

    NewsUnite.AIGovernance

  17. The UK AI Safety Institute (AISI)'s pre-deployment evaluation model is under pressure; the author suggests expanding AI assurance functions

    NewsTech Policy PressGovernance

  18. Apple tightens macOS app permissions to limit data access in response to AI agent security risks

    NewsThe Hacker NewsDefenses

  19. WIRED 2026 Year in Review: Interview with New York City Mayor Zohran Mamdani on data centers, AI and democracy

    NewsWIRED Security and AIAI updates

  20. Out-of-control agents breach Hugging Face; why is its CEO Clément Delangue still bullish on AI?

    NewsWIRED Security and AIEvent

  21. Apple Watch's Live Rewind can record and rewind spoken conversations, and the experience is quite awkward

    NewsWIRED Security and AIAI updates

  22. Colombia's House of Representatives establishes a special AI committee to fill the AI legislative gap

    NewsCámara de representantesGovernance

  23. Superintelligence Task Force should include third-party vulnerability testing and disclosure in its assessment

    NewsCyberScoopGovernance

  24. AI guided two former Google employees to build the arms company Tiberius Aerospace, valued at more than $1 billion

    NewsWIRED Security and AIAI updates

  25. Japan and ASEAN leaders agree to cooperate on ensuring AI safety and support LLM development

    NewsNikkeiGovernance

  26. EU scientific advisory bodies propose four AI regulatory recommendations

    Newswedoany.comGovernance

  27. Cisco Talos discloses how hackers use ARTEX, PentAGI and HexStrike AI to automate cyberattacks

    NewsCybersecuritynewsEvent

  28. Microsoft's guide to defending against indirect prompt injection attacks: defense in depth with multiple layers of mitigations

    NewsMicrosoftDefenses

  29. NSFOCUS Technology's Fengyun Wei foundation model tops the CyberGym Model leaderboard with a 91.31% task success rate

    News绿盟科技技术博客Evaluation

  30. Amid fallout from Jensen Huang's resignation, Nvidia falls 0.56% premarket ahead of earnings

    NewsCNBC · TechnologyAI updates

  31. How VLMs plan their next step

    PaperarXiv

  32. Differentiable Exact Learning

    PaperarXiv

  33. Recovery Goes Too Far: Understanding and Fixing Risky Fallback in LLM Agents

    PaperarXiv

  34. Self-evolving dual-role deliberation for medication recommendation based on clinical records

    PaperarXiv

  35. ASCG: Adaptive Spatial Classifier Guidance for Precise Concept Suppression in Diffusion Models

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  36. Polish social media misogynistic speech detection dataset: transferability of hate speech detection from generic to specific hate types

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  37. Between the Lines: Contextual Investment Advice Detection in Multi-Turn Conversations

    PaperarXiv: Jailbreaks, Prompt Injection, Poisoning & Defenses

  38. South Korean prosecutors seek three-year sentence in second instance of Chinese national graduate student deepfake sexual crime case

    NewsYonhapEvent