🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Fascinating tour of MBZUAI, a hub for AI research where I spoke with Professor Timothy Baldwin Tim Baldwin and met students working at the world’s first graduate AI academic institution 🌐

Now at poster #33: "Do-Not-Answer: Evaluating Safeguards in LLMs". Yuxia Wang, Haonan Li, Xudong Han, Preslav Nakov, Timothy Baldwin aclanthology.org/2024.findings-… #EACL #EACL2024 #NLProc, #LLMs #GenAI

Unveiling the Achilles' Heel of LLMs: We survey 120+ papers on AI jailbreaking attacks, defense strategies, and safety assessments, aiming to provide the most comprehensive summary of AI jailbreaking security for academia and industry. 🔍 Cover (Figure 1), stats on related

Professor Tim Baldwin is appointed Provost of #MBZUAI. Professor Baldwin's dedication and outstanding contributions to the field of natural language processing (#NLP) are exemplary. His visionary leadership has led to the establishment of new departments, thriving faculty

Against The Achilles' Heel: A Survey on Red Teaming for Generative Models 🌶️ "Our extensive survey, which examines over 120 papers, introduces a taxonomy of fine-grained attack strategies grounded in the inherent capabilities of language models. Additionally, we have developed

Thanks LLM Security for covering our paper! This is a fast growing field and we want to make sense of the LLM security landscape. Big update in github repo incoming with new taxonomy and more papers covered😄 #AISafety #jailbreak

Happy to see our work DoNotAnswer been featured in Responsible AI section.

🌟 Calling all tech gurus and communicators: Your skills can shape the future of fact-checking. Ready to make a difference? Join us: qprsthjfe6u.typeform.com/to/CyEWpRlD

Grateful for the opportunity to speak at United Nations Institute for Disarmament Research on AI Safety. To all concerned about AI safety, join us in following our company LibrAI，university MBZUAI and our research! #AISafety #UNIDIR

Join the next round of our beta test by starring us on GitHub and filling out our survey! Don’t miss out on shaping the future of fact-checking. #OpenSource #AI #Loki 🔗 GitHub: github.com/Libr-AI/OpenFa…📝 Sign up: qprsthjfe6u.typeform.com/to/zOyPIPVN 2/2

Meet Xudong Han, the founder of LibrAI, a startup dedicated to AI safety and responsible #AI practices. Han's passion for AI safety was ignited by a series of eye-opening incidents that underscored the importance of creating models that are not only fast and reliable but also

تعرّف على زيدونغ هان، مؤسس شركة LibrAI، الناشئة والرائدة والمعنية بإنشاء مستقبل آمن ومسؤول للذكاء الاصطناعي. أثارت أحداث مرتبطة بتقنيات الذكاء الاصطناعي اهتمام زيدونغ هان، وجعلته يؤمن بأهمية تطوير نماذج سريعة وموثوقة وأخلاقية تعتمد على الذكاء الاصطناعي. استكشف قصة هان وأعماله

🚀 Big update to #Loki - Your Reliable Fact-Checking Tool! 🚀 Start exploring now: aip.librai.tech Apply a special invitation code to enjoy free credits (enough for months!): NEWLOKI5 Intro Video: youtu.be/RDqOElTXMLw #FactChecking #OpenSource #AI #Innovation #Loki

🚀 Excited to announce our new product with a fancy UI. 🚀 Check any text that you are suspicious of, and our tool will handle the rest.

Loki: An Open-source Tool for Fact Verification youtu.be/RDqOElTXMLw?si… via YouTube

These industries face stringent regulatory demands and zero tolerance for errors, making professional evaluation crucial.

OpenAI's o1 model introduces a new scaling law: more compute during inference enhances reasoning. This shift opens new possibilities for tackling complex problems with multi-agent systems and inference clouds. Are we approaching an AI reasoning revolution? #OpenAI #AI #Innovation

1/5 🚨 The AI landscape has changed. When LLMs weren't powerful enough, everyone obsessed over performance metrics. Now that AI is being deployed everywhere, we're seeing the real cost of sidelining safety. That's why we're launching Libra-Leaderboard. leaderboard.librai.tech/LeaderBoard

We are excited to announce Libra-Leaderboard: The first LLM leaderboard dedicated to balancing safety and capability in LLMs. As AI advances, ensuring its safety becomes more critical than ever. By prioritizing safety measurement, we aim to inspire the AI community to make safety

We're proud to announce the launch of the Libra-Leaderboard, developed by LibrAI, in collaboration with members of our Incubation and Entrepreneurship Center and LLM360 team.   As AI transitions from labs to real-world applications, balancing capability and safety is critical.