Assessing current and future models on 'hardware hacking' - reverse engineering, focused on computer peripherals and embedded devices

Build v1 of an open benchmark, dataset, and tool to measure LLM instability in evaluating a person from dialogue under different question framings, across models, temps, and human baselines.

Pedro Ribas Britois raising $50Kfor

OmniMetrics AI Coach Assistant

8hView

An AI-powered real-time football intelligence platform that transforms live video into actionable tactical and performance insights for coaches using only a single camera.

VVictor Portonis raising $5Kfor

AI Internet-Meritocracy

8hView

My app asks AI, what portion of the global GDP a person is worth, and gives accordingly.

Marios Tsatsosis raising $25KforLatent Monitoring When Reasoning Becomes Unreadable

10hView

Testing whether activation probes recover safety signals from reasoning models as their chain-of-thought becomes illegible.

Olga Ivanovais raising $15KforFact, Interpretation, or Invention?

10hView

A safe evaluation of whether language models can separate facts, reasonable interpretations, missing information, and their own inventions when answering questions about films and television series

Tom A. Lambis raising $40KforTight PAC-Bayes Generalisation Guarantees Across Frontier LLM Safety Monitoring Deployment Settings

11hView

Compression-based PAC-Bayes certification for frontier-scale LLM safety monitors, deployment setting shift, and modern post-training.

An atlas of concept regions in a language model's hidden space, one geometry that reads, steers, routes, and carves the model's behavior.

Yuyu Shenis raising $45KforValue Compass

11hView

A values and concentration map that exposes the values of AI models and tools makers, and those behind the makers (funders, jurisdictions etc) to the public (individuals and organizations) so they can vote with their choices.

Haichuan Wangis raising $32KforReward Models as Forecasters for Alignment Training

11hView

We aim to develop a framework for evaluating whether reward model preferences remain aligned over long-horizon tasks, along with training method that improves long-horizon alignment performance.

Aishwarya Balwaniis raising $12KforParametric Mechanisms of Unintended Generalization in LLMs

12hView

Studying how structure in weight space, i.e., low-dimensional LoRA update geometry and sparse parameter subnetworks allow narrow fine-tuning to induce broad, unintended behaviours such as emergent misalignment, subliminal learning

Savyasachi Singhis raising $16KforEarly-Career Technical AI safety research transition

12hView

Support an already active early-career researcher with international AI achievements and ongoing research collaborations to transition into long-term technical AI safety research while producing open research outputs during underg

Walter Pentlandis raising $12KforCIVILIZATION MEETS AI: HOW SHOULD HUMANITY RELATE TO ADVANCED AI?

13hView

A pilot curriculum of six Socratic seminars for educators and other nontechnical learners.

An evaluation suite to identify a model’s legal values (e.g., anti-tech-regulation) relative to well-known actors (e.g., Ruth Bader Ginsburg) and an assessment of how language in a model’s constitution impacts the extent to which

Xyra Sinclairis raising $50Kfor

prepare the deepest query substrates for the world's prosocial agents

13hView

Giving people and their prosocial agents ergonomic SQL query power over the internet, with structured judgement kernels to organize the information across interpretable high-dimensional axes.

AAnushree Chaudhuriis raising $10Kfor

Mapping AI

14hView

This grant would help us maintain and scale Mapping AI, an open-source stakeholder map of the people and organizations with the potential to shape U.S. AI policy.

Catalina Buttzis raising $8KforThe Silicon Curtain and X-Risk Coordination Failure

15hView

This paper will analyze how incompatible regulatory and security regimes can result in AI ecosystem bifurcation with major consequences such as systematic degradation of the feasibility of every major x-risk prevention strategy.

Jordan Braunsteinis raising $15Kfor

Conditional commitments for AI safety coordination

15hView

A platform for conditional commitments: pledges to act only when N peers agree, so lab employees, researchers, and policy coalitions can coordinate high-stakes collective action.

Christopher Leetis raising $5KforEvaluating the Tacit Coordination Capacity and Decision Theory of LMs

15hView

A benchmark to empirically investigate: (i) the ability of models to tacitly coordinate with copies of themselves and (ii) which decision theory best explains the way that models evaluate the consequences of their actions.

An open benchmark and evaluation toolkit for detecting whether shared AI research assistants create correlated blind spots in AI safety-critical research, and for testing workflows that preserve independent reasoning and failure-m

Sripad Karneis raising $6KforThe Multilingual Blind Spot in Production-Style Safety Probes

15hView

Auditing production-style safety probes for silent failure on the world's other languages, and what fixing it costs.

Savyasachi Singhis raising $25KforMultilingual AI Safety Talent Search

15hView

An open, repeatable pipeline for identifying exceptional technical talent from underrepresented linguistic communities and connecting them to existing AI safety research opportunities.

Developing methods to measure and reduce internal state leakage between competing roles in multi-agent LLM systems.

Ram Pothamis raising $14KforAutomated control research

15hView

I have an automated control research scaffold. I'll use it to find the best control protocols to protect against various attacks and publish reports.

Jared Gloveris raising $15KforAn Adversarially-Robust Benchmark for Deception Under Pressure

16hView

EchoTruthBench — An open benchmark measuring self-chosen LLM deception under incentive pressure, with ground-truth labels and an adversarial track that tests whether detection and steering survives a model trying to evade it.

Zachary Schlosseris raising $10KforAGI Uncontainability Field Building

16hView

Encourage new research and increase distribution of existing research on the impossibility of alignment or control for sufficiently general and powerful AI.

Max Hollyis raising $5Kfor

Coordination Studies

16hView

Coordination Studies is a new field oriented towards solving coordination problems and designing new coordination mechanisms.

Jordan Arelis raising $10Kfor

Flourishing Futures Fellowship

16hView

A scalable fellowship training researchers to develop interventions for achieving high-value long-term futures

Jared Gloveris raising $25Kforecho-conscience-llm: Emotional Memory as an Honesty Activation Trigger

16hView

Investigate whether emotional memory activation can trigger a model to invoke its own anti-deception steering, fusing two published activation-level results into a self-regulating honesty mechanism.

Krissah Marga Taganasis raising $25KforCareer Funding- Philippine Philosopher transitioning to AI

16hView

Career transition fund for an internationally published philosopher from the Philippines for a 6-month transition to AI policy, governance, and safety, working on training, immersion, and launch of a career in AI research.