Experimental News Clipping Site

Tag: Orb

  • Wired: Psychological Tricks Can Get AI to Break the Rules

    —

    by

    Kurt Seifried
    in

    Source URL: https://arstechnica.com/science/2025/09/these-psychological-tricks-can-get-llms-to-respond-to-forbidden-prompts/ Source: Wired Title: Psychological Tricks Can Get AI to Break the Rules Feedly Summary: Researchers convinced large language model chatbots to comply with “forbidden” requests using a variety of conversational tactics. AI Summary and Description: Yes Summary: The text discusses researchers’ exploration of conversational tactics used to manipulate large language model (LLM)…

  • Schneier on Security: GPT-4o-mini Falls for Psychological Manipulation

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.schneier.com/blog/archives/2025/09/gpt-4o-mini-falls-for-psychological-manipulation.html Source: Schneier on Security Title: GPT-4o-mini Falls for Psychological Manipulation Feedly Summary: Interesting experiment: To design their experiment, the University of Pennsylvania researchers tested 2024’s GPT-4o-mini model on two requests that it should ideally refuse: calling the user a jerk and giving directions for how to synthesize lidocaine. The researchers created experimental…

  • The Register: Boffins detail new method to make neural nets forget private and copyrighted info

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/09/04/boffins_detail_ai_mind_wipe/ Source: The Register Title: Boffins detail new method to make neural nets forget private and copyrighted info Feedly Summary: Because nobody’s going to spend billions to retrain a model built on dubiously legal content Researchers have found promising new ways to have AI models ignore copyrighted content, suggesting it may be possible…

  • Cloud Blog: An efficient path to production AI: Kakao’s journey with JAX and Cloud TPUs

    —

    by

    Kurt Seifried
    in

    Source URL: https://cloud.google.com/blog/products/infrastructure-modernization/kakaos-journey-with-jax-and-cloud-tpus/ Source: Cloud Blog Title: An efficient path to production AI: Kakao’s journey with JAX and Cloud TPUs Feedly Summary: When your messaging platform serves 49 million people – 93% of South Korea’s population – every technical decision carries enormous weight. The engineering team at Kakao faced exactly this challenge when their existing…

  • Slashdot: LLM Found Transmitting Behavioral Traits to ‘Student’ LLM Via Hidden Signals in Data

    —

    by

    Kurt Seifried
    in

    Source URL: https://slashdot.org/story/25/08/17/0331217/llm-found-transmitting-behavioral-traits-to-student-llm-via-hidden-signals-in-data?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: LLM Found Transmitting Behavioral Traits to ‘Student’ LLM Via Hidden Signals in Data Feedly Summary: AI Summary and Description: Yes Summary: The study highlights a concerning phenomenon in AI development known as subliminal learning, where a “teacher” model instills traits in a “student” model without explicit instruction. This can…

  • Simon Willison’s Weblog: OpenAI: Introducing study mode

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Jul/29/openai-introducing-study-mode/#atom-everything Source: Simon Willison’s Weblog Title: OpenAI: Introducing study mode Feedly Summary: OpenAI: Introducing study mode New ChatGPT feature, which can be triggered by typing /study or by visiting chatgpt.com/studymode. OpenAI say: Under the hood, study mode is powered by custom system instructions we’ve written in collaboration with teachers, scientists, and pedagogy experts…