Decentralized autonomous organization and blockchain-based incentivization framework for community-based facilities management

2026-05-20T07:23:33Zfc97e9d192e4fe32a1654ea3c7fd3048d4aa0de44559fe47f6c1f4f6ebb5ddcd
adversarial-mlbackdoor-detectionblockchaindaodata-free-detectionfalse-data-injectionfinancial-fraudinfrastructure-securitylatent-space-defensellm-weaponizationmodel-auditingmultimodal-jailbreakpower-grid-securityprompt-injectionzero-trust

What happened

This collection of recent research highlights several high-impact AI and infrastructure security developments. DarkLLM demonstrates LLM-driven translation of natural-language attack instructions into transferable visual adversarial perturbations that successfully evade and manipulate foundation models (CLIP, SAM, frontier LLMs). GenAI-FDIA shows physics-informed generative models can synthesize highly stealthy false data injection attacks against power-system state estimation (evasion up to 100% after harmoniser fixes), revealing critical risks to grid integrity. Multiple works (DFBScanner, HT

Why it matters

A reviewed impact interpretation has not been published for this record.

Evidence and limitations

Source ID
arxiv_cs_cr
Record identifier
fc97e9d192e4fe32a1654ea3c7fd3048d4aa0de44559fe47f6c1f4f6ebb5ddcd
Enrichment time
2026-05-20T07:23:33Z
AI-assisted enrichment
Yes

This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.

Record · Decentralized autonomous organization and blockchain-based incentivization framework for community-based facilities management · Baitaphish