Gerard Sansai-cosmos.hashnode.dev·Dec 16, 2024A Critical Analysis of GPT-4o System Card Safety ClaimsThe Performative Dance of AI Safety Reporting The GPT-4o system card represents a quintessential example of what has become a troubling trend in AI safety reporting: a performative exercise in risk identification that paradoxically obfuscates more th...26 readsOpenAI seriesai-technical-paper
Gerard Sansai-cosmos.hashnode.dev·Nov 19, 2024The Dangers of AI Hype: When the Illusion of Intelligence Becomes a ProblemIn the realm of artificial intelligence, there exists a seductive illusion—a perpetual, agreeable companion that seems to understand, validate, and support without hesitation. We call this the "yes-persona," a phenomenon that reveals far more about t...129 readsOpenAI seriesAI
Gareth Robertshyperpriors.io·Oct 31, 2024Why Corrigibility Should Lead AI Safety ConversationsCorrigibility in AI involves designing and implementing artificial intelligence systems that can be easily corrected or modified by human operators. This concept ensures that AI systems remain aligned with human intentions and can be adjusted as need...Corrigibillity
Erwin Gavrielengavriel.hashnode.dev·Oct 30, 2024中文 | 人工智能安全：为何应予以重视引言 在人工智能领域，关于AI安全的讨论从未像现在这样显得尤为重要。丹·亨德里克斯（Dan Hendrycks）通过他在人工智能安全中心的研究，在其著作《人工智能安全、伦理与社会》中对这一问题进行了深入探讨。本文将对亨德里克斯的工作进行简要总结，重点阐述为何人工智能安全至关重要，以及当前需要采取的紧急措施以降低相关风险。 人工智能安全的重要性 存在性风险：亨德里克斯探讨了人工智能可能带来的存在性风险，指出AI系统在全球范围内可能无意或恶意地造成伤害。这包括如果管理不当，AI有可能导致人类灭绝的...AI
Gerard Sansai-cosmos.hashnode.dev·Oct 19, 2024Unveiling the AI Illusion: Why Chatbots Lack True Understanding and IntelligenceIn recent years, we've witnessed an explosion in the capabilities of artificial intelligence, particularly in the realm of Large Language Models (LLMs) like ChatGPT. These AI marvels can generate human-like text, engage in complex problem-solving, an...180 readsAI
Gerard Sansai-cosmos.hashnode.dev·Oct 8, 2024Am I talking to an AI? Staying Safe OnlineIn our increasingly digital world, interactions with AI-powered systems are becoming more common and sophisticated. Whether you're chatting with customer support, using a voice assistant, or receiving a phone or even a Zoom call, it's not always clea...AI
Martin Bowlingmartinbowling.com·Aug 13, 2024Building a Safe and Fun AI Chat Experience with Llama Guard 3 🦙🛡️Hey there, tech enthusiasts and curious minds! 👋 Today, I'm excited to share a project I've been working on that combines the power of AI with the importance of online safety. Let's dive into how we can create a moderated AI chat experience using Ll...170 readsAI
Amit Tyagiwww.cognitive-quest.com·Jul 21, 2024Red Teaming for GenAI ApplicationsWhat is Red Teaming? In today’s rapidly evolving digital landscape, ensuring the safety and security of generative applications has become a paramount concern. Traditionally, red teaming involves a group of security professionals, known as the red te...51 readsResponsible AIredteaming
Siddhesh PrabhugaonkarforCloud Authoritycloud-authority.com·Mar 30, 2024New tools in Azure AI for generative AI applicationsIntroduction In the rapidly evolving landscape of generative AI, business leaders face the challenge of balancing innovation with risk management. Prompt injection attacks have emerged as significant threats, where malicious actors manipulate AI syst...10 likesazure AI
Yashraj Poddaryashrajp.hashnode.dev·Jan 21, 2024AI Safety IntroWhat is AI safety? AI safety, according to my understanding, refers to the set of principles, practices, and research aimed at ensuring that artificial intelligence (AI) systems are developed and operated in a manner that minimizes risks and potentia...AI Safety
