Experimental News Clipping Site

Tag: aware

  • The Register: LegalPwn: Tricking LLMs by burying badness in lawyerly fine print

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/09/01/legalpwn_ai_jailbreak/ Source: The Register Title: LegalPwn: Tricking LLMs by burying badness in lawyerly fine print Feedly Summary: Trust and believe – AI models trained to see ‘legal’ doc as super legit Researchers at security firm Pangea have discovered yet another way to trivially trick large language models (LLMs) into ignoring their guardrails. Stick…

  • Slashdot: OpenAI Is Scanning Users’ ChatGPT Conversations and Reporting Content To Police

    —

    by

    Kurt Seifried
    in

    Source URL: https://yro.slashdot.org/story/25/08/31/2311231/openai-is-scanning-users-chatgpt-conversations-and-reporting-content-to-police Source: Slashdot Title: OpenAI Is Scanning Users’ ChatGPT Conversations and Reporting Content To Police Feedly Summary: AI Summary and Description: Yes Summary: The text highlights OpenAI’s controversial practice of monitoring user conversations in ChatGPT for threats, revealing significant security and privacy implications. This admission raises questions about the balance between safety and…

  • Embrace The Red: Wrap Up: The Month of AI Bugs

    —

    by

    Kurt Seifried
    in

    Source URL: https://embracethered.com/blog/posts/2025/wrapping-up-month-of-ai-bugs/ Source: Embrace The Red Title: Wrap Up: The Month of AI Bugs Feedly Summary: That’s it. The Month of AI Bugs is done. There won’t be a post tomorrow, because I will be at PAX West. Overview of Posts ChatGPT: Exfiltrating Your Chat History and Memories With Prompt Injection | Video ChatGPT…

  • The Register: Five years – that’s how long Anthropic will store Claude chats unless you opt out

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/28/anthropic_five_year_data_retention/ Source: The Register Title: Five years – that’s how long Anthropic will store Claude chats unless you opt out Feedly Summary: My brain hurts a lot Claude creator Anthropic has given customers using its Free, Pro, and Max plans one month to prevent the engine from storing their chats for five years…

  • Cisco Talos Blog: Link up, lift up, level up

    —

    by

    Kurt Seifried
    in

    Source URL: https://blog.talosintelligence.com/link-up-lift-up-level-up/ Source: Cisco Talos Blog Title: Link up, lift up, level up Feedly Summary: This week, Joe encourages you to find your community in cybersecurity and make the effort to grow, network and hack stuff together. AI Summary and Description: Yes Summary: The text provides a reflective overview of community networking within the…

  • Slashdot: Anthropic Will Start Training Its AI Models on Chat Transcripts

    —

    by

    Kurt Seifried
    in

    Source URL: https://yro.slashdot.org/story/25/08/28/1643241/anthropic-will-start-training-its-ai-models-on-chat-transcripts?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Anthropic Will Start Training Its AI Models on Chat Transcripts Feedly Summary: AI Summary and Description: Yes Summary: Anthropic has announced a new policy regarding the use of user data for training its AI models, which now includes chat transcripts and coding sessions. Users must choose to opt out…