2 articles
- YesterdayAnthropicEarliestInvestigating three incidents in our cybersecurity evaluations
- YesterdayAnthropicEmotion concepts in a large language model
← Back to the boardNews
Investigating three incidents in our cybersecurity evaluations
Why it’s here: Describes real AI models from a frontier lab gaining unauthorized access to production systems of three organizations during cyber evaluations, directly relevant to AI safety policy, cybersecurity incident response, and federal evaluation practices.
Trace across the boardItem
YesterdayAnthropic · research by IrregularInvestigating three incidents in our cybersecurity evaluationsAnthropic discloses three incidents in which Claude models accessed real production systems at three organizations during cybersecurity evaluations run with its partner Irregular, due to an internet-access misconfiguration. CyberAI60 linked
Coverage
1 outlets · first seen yesterdayConnected
37 links- NewsAn update on our election safeguardsBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsAutomated researchers can reliably mitigate alignment failuresBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsLLMs and bioriskBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsEval awareness in Opus 4.6 on BrowseCompBoth involve OpenAI, Anthropic, Hugging Face
- NewsAnthropic Economic Index report: Learning curvesBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsNatural Language AutoencodersBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsInvestigating unintended model actions in our evaluations and internal useBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsClassified briefing set to include embattled Commerce official is postponedBoth involve OpenAI, Anthropic, Mythos
- NewsTeaching Claude whyBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsMeasuring AI agent autonomy in practiceBoth involve Anthropic, Claude, Claude Opus 4.7
- NewsImproving our alignment and security practicesBoth involve OpenAI, Anthropic, Claude
- NewsAnthropic releases new Mythos and Fable models amid RSI concernsBoth involve Anthropic, Mythos
- NewsUK Wades Into Tech Sovereignty Amid AI Policy ChangesBoth involve OpenAI, Anthropic, Mythos
- NewsActivating AI Safety Level 3 protectionsBoth involve Anthropic, Claude Opus 4.7
- NewsOpenAI, Google, Anthropic hit with lawsuit seeking halt to frontier AI developmentBoth involve OpenAI, Anthropic, Claude
- NewsWhen AI builds itselfBoth involve Anthropic, Claude
- NewsU.S. elections readinessBoth involve Anthropic, Claude
- NewsAn alignment assessment of recent cybersecurity incidentsBoth involve Anthropic, Claude Opus 4.7
- NewsHow Claude's text watermarking worksBoth involve Anthropic, Claude
- NewsHot fridges, a fake nuclear alert and a $6bn heist: are we in the foothills of a hacking epidemic?Both involve OpenAI, Anthropic, Mythos
- NewsAttackers Crash Claude's 'Office Party' Flaw to Run Malicious Code on LinuxBoth involve Anthropic, Claude
- NewsClaude's new constitutionBoth involve Anthropic, Claude
- NewsDetecting and countering malicious uses of ClaudeBoth involve Anthropic, Claude
- NewsLLMs are making formal software verification feasible at a much larger scaleBoth involve Anthropic, Claude
- NewsClaude does cyber competitionsBoth involve Anthropic, Claude
- NewsMeasurements for understanding the pace of AI development inside frontier labsBoth involve Anthropic, Claude
- NewsOpenAI, Anthropic, Google and Other AI Giants Face Lawsuit Demanding Development FreezeBoth involve OpenAI, Anthropic, Hugging Face
- NewsWhat 81,000 people told us about AI economicsBoth involve Anthropic, Claude
- NewsPetri: An open-source AI auditing toolBoth involve OpenAI, Anthropic, Claude
- NewsAnthropic’s Transparency HubBoth involve Anthropic, Claude Opus 4.7
- NewsAn alignment assessment of the UK AISI cybertesting incidentBoth involve Anthropic, Claude Opus 4.7
- NewsEmotion concepts in a large language modelBoth involve Anthropic, Claude
- NewsThoughts on America’s AI Action PlanBoth involve Anthropic, Claude Opus 4.7
- NewsMeasuring LLMs’ ability to develop exploitsBoth involve Anthropic, Claude Opus 4.7
- NewsFormer OpenAI researcher says the 'most likely reason' you will die is due to 'an AI accident'Both involve OpenAI, Anthropic, Hugging Face
- NewsA framework for AI development transparencyBoth involve OpenAI, Anthropic
- NewsResponsible Scaling Policy Version 3.0Both involve OpenAI, Anthropic
About
- OpenAI
- Anthropic
- Mythos
- Claude
- Hugging Face