1/26/2024 0 Comments Defcon schedule 2022AIV hosted the first public bias bounty at DEF CON 29 which has grown into the Bias Buccaneers and Humane Intelligence and we’re working with that team again on this event. The DEF CON community has extensive experience evaluating a huge range of technologies. In addition to thousands of hackers, we are also bringing in partners from community groups and policy-oriented nonprofits as well as supporters in government. The individual who gets the highest number of points wins a high end NVIDIA GPU. Red teamers will be expected to abide by the hacker hippocratic oath. We will also be providing a capture the flag (CTF) style point system to promote testing a wide range of harms. We will be providing laptops and timed access to multiple LLMs from the vendors. This is also an opportunity for new communities to learn skills in AI by exploring its quirks and limitations. The more people who know how to best work with these models, and their limitations, the better. We’ll publish what we learn from this event to help others who want to try the same thing. This is the first time anyone is attempting more than a few hundred experts to assess these models, so we will be learning together. Thousands of people will experience hands-on LLM red-teaming for the first time – and we’re bringing in hundreds of students from overlooked institutions and communities. These fill two needs with one deed, addressing the harms and growing the community of researchers that know how to help.”Īt DEF CON 2023, we are conducting the largest red teaming exercise ever for any group of AI models. Bug bounties, live hacking events, and other standard community engagements in security can be modified for machine learning model based systems. The diverse issues with these models will not be resolved until more people know how to red team and assess them. However this work has largely happened in private. Hallucinations, jailbreaks, bias, and a drastic leap in capabilities are all new issues security professionals and the public have to deal with.Īccording to Sven Cattell, the founder of AI Village, “Traditionally, companies have solved this problem with specialized red teams. However, we’re only beginning to understand the embedded and emergent risks that come from automating these new technologies at scale. They can lower barriers to entry in creative fields, and allow for new kinds of creative content. They can help people get their ideas out faster and better than ever before. We love the explosion of creativity that new generative large language models (LLMs) allow. We will be testing models kindly provided by Anthropic, Google, Hugging Face, NVIDIA, OpenAI, and Stability with participation from Microsoft, on an evaluation platform developed by Scale AI. This event is supported by the White House Office of Science, Technology, and Policy, the National Science Foundation’s Computer and Information Science and Engineering (CISE) Directorate, and the Congressional AI Caucus.ĪI Village (AIV) is hosting the first public generative AI red team event at DEF CON 31 with our partners at Humane Intelligence, SeedAI, and the AI Vulnerability Database. Largest annual hacker convention to host thousands to find bugs in large language models built by Anthropic, Google, Hugging Face, NVIDIA, OpenAI, and Stability. Posted by Sven Cattell, Rumman Chowdhury, Austin Carson on ![]() ![]() ![]() AI Village at DEF CON announces largest-ever public Generative AI Red Team
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |