Experimental News Clipping Site

Tag: -4o

  • Simon Willison’s Weblog: Hallucinations in code are the least dangerous form of LLM mistakes

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Mar/2/hallucinations-in-code/#atom-everything Source: Simon Willison’s Weblog Title: Hallucinations in code are the least dangerous form of LLM mistakes Feedly Summary: A surprisingly common complaint I see from developers who have tried using LLMs for code is that they encountered a hallucination – usually the LLM inventing a method or even a full software library…

  • Simon Willison’s Weblog: Structured data extraction from unstructured content using LLM schemas

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Feb/28/llm-schemas/#atom-everything Source: Simon Willison’s Weblog Title: Structured data extraction from unstructured content using LLM schemas Feedly Summary: LLM 0.23 is out today, and the signature feature is support for schemas – a new way of providing structured output from a model that matches a specification provided by the user. I’ve also upgraded both…

  • OpenAI : Supporting sellers with enhanced product listings

    —

    by

    Kurt Seifried
    in

    Source URL: https://openai.com/index/mercari Source: OpenAI Title: Supporting sellers with enhanced product listings Feedly Summary: Mercari leverages GPT-4o mini and GPT-4 to streamline selling, enhance product listings, and boost sales, transforming the online marketplace with features like AI Listing Support and Mercari AI Assistant. AI Summary and Description: Yes Summary: The text highlights how Mercari is…

  • Simon Willison’s Weblog: Introducing GPT-4.5

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Feb/27/introducing-gpt-45/#atom-everything Source: Simon Willison’s Weblog Title: Introducing GPT-4.5 Feedly Summary: Introducing GPT-4.5 GPT-4.5 is out today as a “research preview" – it’s available to OpenAI Pro ($200/month) customers but also to developers with an API key. OpenAI also published a GPT-4.5 system card. I’ve started work adding it to LLM but I don’t…

  • Schneier on Security: “Emergent Misalignment” in LLMs

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.schneier.com/blog/archives/2025/02/emergent-misalignment-in-llms.html Source: Schneier on Security Title: “Emergent Misalignment” in LLMs Feedly Summary: Interesting research: “Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs“: Abstract: We present a surprising result regarding LLMs and alignment. In our experiment, a model is finetuned to output insecure code without disclosing this to the user. The resulting model…

  • The Register: Does terrible code drive you mad? Wait until you see what it does to OpenAI’s GPT-4o

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/02/27/llm_emergent_misalignment_study/ Source: The Register Title: Does terrible code drive you mad? Wait until you see what it does to OpenAI’s GPT-4o Feedly Summary: Model was fine-tuned to write vulnerable software – then suggested enslaving humanity Computer scientists have found that fine-tuning notionally safe large language models to do one thing badly can negatively…

  • Slashdot: Inception Emerges From Stealth With a New Type of AI Model

    —

    by

    Kurt Seifried
    in

    Source URL: https://slashdot.org/story/25/02/26/2257224/inception-emerges-from-stealth-with-a-new-type-of-ai-model?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Inception Emerges From Stealth With a New Type of AI Model Feedly Summary: AI Summary and Description: Yes Summary: Inception, a startup led by Stanford professor Stefano Ermon, has developed a highly efficient diffusion-based large language model (DLM) that surpasses traditional models in both speed and cost-effectiveness. By enabling…

  • Simon Willison’s Weblog: olmOCR

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Feb/26/olmocr/#atom-everything Source: Simon Willison’s Weblog Title: olmOCR Feedly Summary: olmOCR New from Ai2 – olmOCR is “an open-source tool designed for high-throughput conversion of PDFs and other documents into plain text while preserving natural reading order". At its core is allenai/olmOCR-7B-0225-preview, a Qwen2-VL-7B-Instruct variant trained on ~250,000 pages of diverse PDF content (both…

  • Simon Willison’s Weblog: Quoting Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Feb/25/emergent-misalignment/ Source: Simon Willison’s Weblog Title: Quoting Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs Feedly Summary: In our experiment, a model is finetuned to output insecure code without disclosing this to the user. The resulting model acts misaligned on a broad range of prompts that are unrelated to coding: it asserts…

  • Hacker News: Narrow finetuning can produce broadly misaligned LLM [pdf]

    —

    by

    Kurt Seifried
    in

    Source URL: https://martins1612.github.io/emergent_misalignment_betley.pdf Source: Hacker News Title: Narrow finetuning can produce broadly misaligned LLM [pdf] Feedly Summary: Comments AI Summary and Description: Yes **Summary:** The document presents findings on the phenomenon of “emergent misalignment” in large language models (LLMs) like GPT-4o when finetuned on specific narrow tasks, particularly the creation of insecure code. The results…