Decentralized Assessment for Trustworthy AI (DATA)
Summary
The Decentralized Assessment for Trustworthy AI (DATA) is an ethical evaluation tool that allows users and communities to objectively audit AI companies based on leading ethical frameworks like UNESCO and EU guidelines.
Similar Articles
I trust-scored 171 open-source AI agents — most can't prove their supply chain
A developer created an independent trust registry for 171 open-source AI agents, scoring them on verifiable trust signals like supply chain security and maintenance, finding that only three agents achieved a Grade A rating while many popular agents lacked basic verification.
DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents
This paper introduces the DecodingTrust-Agent Platform (DTap), a controllable and interactive red-teaming platform for evaluating AI agent security across multiple domains. It also presents DTap-Red, an autonomous agent for discovering attack strategies, and DTap-Bench, a large-scale dataset for risk assessment.
What would actually make you trust an AI? Not "it sounds right," but trust it the way you trust a person or an institution?
A discussion exploring what specific conditions (transparency, verifiable track record, persistent identity, accountability) would make people trust AI systems as they trust humans or institutions, rather than just accepting them as tools.
Verifiable Agentic Infrastructure: Proof-Derived Authorization for Sovereign AI Systems
This paper introduces a Distributed Trust Framework (DTF) for verifiable, proof-derived authorization in autonomous AI agent systems, addressing the risks of identity-centric permissions by requiring justification proofs and consensus for execution.
We read the ToS & Privacy Policy for 205 AI apps and graded them. Over half got a D or F.
A study graded 205 AI apps on their data governance practices, finding that over half received a D or F grade, with many apps not disclosing whether user input trains their models.