A few weeks ago, I watched a small team of artificial intelligence agents spend roughly 10 minutes trying to hack into my brand new vibe-coded website. The AI agents, developed by startup RunSybil, worked together to probe my poor site to identify weak spots. An orchestrator agent, called Sybil, oversees several more specialized agents all powered by a combination of custom language models and off-the-shelf APIs. Whereas conventional vulnerability scanners probe for specific known problems, Sybil is able to operate at a higher level, using artificial intuition to figure out weaknesses. It might, for example, work out that a guest user has privileged access--something a regular scanner might miss--and use this to build an attack.

Multi-step reasoning is widely adopted in the community to explore the better performance of language models (LMs). We report on the systematic strategy that LMs use in this process. Our controlled experiments reveal that LMs rely more heavily on heuristics, such as lexical overlap, in the earlier stages of reasoning Figure 1: Illustration of the systematic strategy we discovered when more steps are required to reach an in language models (LMs). When the goal is answer. Conversely, as LMs progress closer distant from the current state in a multi-step reasoning to the final answer, their reliance on heuristics process, the models tend to rely on heuristics, such as decreases. This suggests that LMs track only superficial overlap, which can lead them in the wrong a limited number of future steps and dynamically direction. In contrast, when the goal is within a limited combine heuristic strategies with logical distance, the models are more likely to take rational actions ones in tasks involving multi-step reasoning.

This research examines the polycentric governance of digital assets in blockchain-based Decentralized Autonomous Organizations (DAOs). It offers a theoretical framework and addresses a critical challenge facing decentralized governance by developing a method to identify sybils, or spurious identities. Sybils pose significant organizational sustainability threats to DAOs and other, commons-based online communities, and threat models are identified. The experimental method uses graph deep learning techniques to identify sybil activity in a DAO governance dataset (snapshot.org). Specifically, a Graph Convolutional Neural Network (GCNN) learned voting behaviours and a fast k-means vector clustering algorithm (FAISS) used high-dimensional embeddings to identify similar nodes in a graph. The results reveal that deep learning can effectively identify sybils, reducing the voting graph by 2-5%. This research underscores the importance of sybil resistance in DAOs and offers a novel perspective on decentralized governance, informing future policy, regulation, and governance practices.

An unseen, all-powerful female character who can be trusted to impart divine knowledge about the past, present, and future. Today, her name has been mined from antiquity and attached to a developing artificial intelligence tool for lung cancer risk assessment. LDCT scans of the lungs are currently the most common method by which patients are screened in hopes of finding lung cancer early. Taking screening a step further, SYBIL will analyze her LDCT image data without the help of a radiologist to predict the patient's risk of developing lung cancer within her six years. The researchers showed that over a 6-year period, sybil received c-indexes of 0.75, 0.81, and 0.80 from different sets of lung ldct scans obtained from nlst, mgh, and cgmh, respectively. It is considered good and above 0.8 is considered strong.

The name Sybil has its origins in the oracles of Ancient Greece, also known as sibyls: feminine figures who were relied upon to relay divine knowledge of the unseen and the omnipotent past, present, and future. Now, the name has been excavated from antiquity and bestowed on an artificial intelligence tool for lung cancer risk assessment being developed by researchers at MIT's Abdul Latif Jameel Clinic for Machine Learning in Health, Mass General Cancer Center (MGCC), and Chang Gung Memorial Hospital (CGMH). Lung cancer is the No. 1 deadliest cancer in the world, resulting in 1.7 million deaths worldwide in 2020, killing more people than the next three deadliest cancers combined. "It's the biggest cancer killer because it's relatively common and relatively hard to treat, especially once it has reached an advanced stage," says Florian Fintelmann, MGCC thoracic interventional radiologist and coauthor on the new work. "In this case, it's important to know that if you detect lung cancer early, the long-term outcome is significantly better. Your five-year survival rate is closer to 70 percent, whereas if you detect it when it's advanced, the five-year survival rate is just short of 10 percent."

Lung cancer is the leading cause of cancer death in the United States and around the world. Low-dose chest computed tomography (LDCT) is recommended to screen people between 50 and 80 years of age with a significant history of smoking, or who currently smoke. Lung cancer screening with LDCT has been shown to reduce death from lung cancer by up to 24 percent. But as rates of lung cancer climb among non-smokers, new strategies are needed to screen and accurately predict lung cancer risk across a wider population. A study led by investigators from the Mass General Cancer Center, a member of Mass General Brigham, in collaboration with researchers at the Massachusetts Institute of Technology (MIT), developed and tested an artificial intelligence tool known as Sybil.

Federated Learning has been issued as an extraordinary model for preserving private data. It makes the machine learning model in the global model train without sharing personal and private data which is distributed to many other local devices. However, it is also unable to access the data and analyze it to tell which local data is poisoned[1]. With this disadvantage, Federated learning is easily exposed to Sybil attacks such as transfers of faulty results to the global model. FoolsGold[2], those key is that Sybil's training directions are more similar to other Sybil's than typical average local's directions, is known as one of the best algorithms for defense from Sybil attacks.

Voting rules may fail to implement the will of the society when only some voters actively participate, and/or in the presence of sybil (fake or duplicate) voters. Here we aim to address social choice in the presence of sybils and voter abstention. To do so we assume the status-quo (Reality) as an ever-present distinguished alternative, and study Reality Enforcing voting rules, which add virtual votes in support of the status-quo. We measure the tradeoff between safety and liveness (the ability of active honest voters to maintain/change the status-quo, respectively) in a variety of domains, and show that the Reality Enforcing voting rule is optimal in this respect.

While most of humanity is suddenly on the net, the value of this singularity is hampered by the lack of credible digital identities: Social networking, person-to-person transactions, democratic conduct, cooperation and philanthropy are all hampered by the profound presence of fake identities, as illustrated by Facebook's removal of 5.4Bn fake accounts since the beginning of 2019. Here, we introduce the fundamental notion of a \emph{genuine personal identifier}---a globally unique and singular identifier of a person---and present a foundation for a decentralized, grassroots, bottom-up process in which every human being may create, own, and protect the privacy of a genuine personal identifier. The solution employs mutual sureties among owners of personal identifiers, resulting in a mutual-surety graph reminiscent of a web-of-trust. Importantly, this approach is designed for a distributed realization, possibly using distributed ledger technology, and does not depend on the use or storage of biometric properties. For the solution to be complete, additional components are needed, notably a mechanism that encourages honest behavior and a sybil-resilient governance system.

Machine learning (ML) over distributed data is relevant to a variety of domains. Existing approaches, such as federated learning, compose the outputs computed by a group of devices at a central aggregator and run multi-round algorithms to generate a globally shared model. Unfortunately, such approaches are susceptible to a variety of attacks, including model poisoning, which is made substantially worse in the presence of sybils. In this paper we first evaluate the vulnerability of federated learning to sybil-based poisoning attacks. We then describe FoolsGold, a novel defense to this problem that identifies poisoning sybils based on the diversity of client contributions in the distributed learning process. Unlike prior work, our system does not assume that the attackers are in the minority, requires no auxiliary information outside of the learning process, and makes fewer assumptions about clients and their data. In our evaluation we show that FoolsGold exceeds the capabilities of existing state of the art approaches to countering ML poisoning attacks. Our results hold for a variety of conditions, including different distributions of data, varying poisoning targets, and various attack strategies.