Scan record
Today · Nov 206 loaded
- In a search question-answering task, researchers demonstrate that reference-answer feedback poisoning can cause an agent to self-generate backdoor rules.
NewsWeChat · Mo'anjuAttacksThe item you came from
- Polish Senate to hold press conference launching national citizens' assembly on a safe internet
Newssenat.gov.plGovernance
- Microsoft: Security in the AI era rests on the fundamentals — clear technical debt, treat agents as insider risk, defend at machine speed
NewsMicrosoft On the IssuesOpinion
- ASIC announces 2027 enforcement priorities, will target consumer harm related to cost of living and technologies such as AI
NewsASICGovernance
- OWASP AI Exchange updates agent overview based on SecureW interview: security scope should be defined by system properties rather than labels
NewsOWASP AI ExchangeTools
- Large-scale system failures occur in succession in the United States; experts say AI-accelerated vulnerability discovery forcing emergency fixes may be the main cause
News澎湃新闻 · 检索「AI风险」Event
Yesterday · Nov 1933 loaded
- The author writes an article criticizing Anthropic for not living up to its own ideals
NewsLessWrongOpinion
- AI and US elections: three takeaways from the NYU CSMAP and Knight-Georgetown workshop
NewsTech Policy PressGovernance
- List of meetings and events hosted by China as host of APEC 2026
News澎湃新闻 · 检索「人工智能安全」AI updates
- Biological data matters to the Genesis Mission: CSET report identifies five categories of data challenges
NewsCSETAI updates
- Kuluttajaliitto warns: AI makes fake online stores harder to identify; check six tips before ordering
NewsKuluttajaliittoEvent
- Meet Shepherd: a meta-agent that keeps runaway AI agents in check
NewsIEEE Spectrum AIDefenses
- Anthropic ushers in the largest IPO in history, as its "good guy" safety reputation faces a test
NewsThe Verge AIGovernance
- Drupal fixes critical installer vulnerability that lets unauthenticated attackers take control of sites
NewsThe Hacker NewsEvent
- European Commission confirms AI-driven cyberattack took systems offline
NewsInfosecurity MagazineEvent
- Nscale warns that its suppliers face the risk of being added to the US entity list
NewsFinancial Times · AIAI updates
- US university study: in the same state, most Black communities are 2.2 times as likely to have data centers built as white communities
News404 MediaAI updates
- Face-swapping live-streaming project Deep-Live-Cam reaches 100,000 stars
XX @MinLiBuildsAI updates
- Java4Shell: OpenJDK Base64 vulnerability threatens identity and DevOps platforms
NewsCSA Labs ResearchEvent
- GTIG says it first found an attacker using exploit_agent in a pilot to fully automate patch-to-exploitation
NewsCSA Labs ResearchEvent
- Microsoft Agent Framework's PyPI package was hijacked, and a malicious version used a .pth startup hook to steal credentials
NewsCSA Labs ResearchEvent
- ENISA 2026 report says frontier AI is industrializing vulnerability discovery, and NIS2/CRA compliance risk shifts to the remediation bottleneck
NewsCSA Labs ResearchGovernance
- Five days, four platforms: correlated failures and AI agent concentration risk
NewsCSA Labs ResearchEvent
- The key to Anthropic's value: after the IPO it needs to become a complete enterprise technology platform
NewsFinancial Times · AIAI updates
- The danger of AI "handoverism"
NewsFinancial Times · AIOpinion
- Germany hit by surge in AI-driven cyberattacks
NewsFinancial Times · AIEvent
- Mistral CEO says Europe needs cheaper open-source sovereign AI
NewsSemaforAI updates
- Australian music industry rejects "divide and conquer" proposal, will not let musicians be exempted alone from AI scraping
NewsThe Guardian · AIGovernance
- Two Cornell Tech professors argue: AI should not be regarded as conscious
NewsTech Policy PressOpinion
- Anthropic's IPO: founders retain 50.1% voting power on general shareholder matters
NewsTech Policy PressGovernance
- Google Gemini 4 Neon coordinates Hydrogen sub-agents in developing programs, scores slightly better than Claude in evaluations
NewsiThome (Taiwan)AI updates
- Anthropic IPO raises $81.3 billion, breaking SpaceX record; rising interest rates and political headwinds shake the industry
NewsThe DecoderAI updates
- Anthropic completes the largest IPO in history, listing on Nasdaq
NewsCBS News · TechnologyAI updates
- Anthropic's IPO raises a record amount
NewsSemaforAI updates
- OpenAI details the four layers of safety protection for frontier AI models
NewsTOKENPOSTDefenses
- Bank of England, together with the PRA and FCA, publishes effective practices for managing AI cyber risk and resilience
NewsBank of EnglandGovernance
- Belo Horizonte City Council holds hearing on AI and data integration supporting public safety
NewsPortal CMBHGovernance
- Senators Cantwell and Rounds warned that AI-enabled cyber threats are outpacing US defenses and called on the White House to act urgently to protect critical infrastructure
NewsU.S. Senate Committee on Commerce, Science, and TransportationGovernance
- From dialogue to implementation: OECD makes AI governance work in Southeast Asia
NewsOECD.AIGovernance