Explore 2 source-supported AI security cases for 1.3 · Poor data quality in Databricks DASF 3.0, with evidence, scope, and defensive analysis.
Risk category
Data used by an AI system lacks sufficient accuracy, completeness, freshness, or consistency with applicable rules. Poor quality may result from malicious changes or ordinary mistakes and can undermine analytics and model-based decisions. Tracing data origins and transformations, applying quality rules, and monitoring changes help identify where reliability has been lost.
Zenity tested an AWS Bedrock AgentCore deployment, a managed environment for running AI agents, with a network-capable agent and a broad execution role. The agent relayed temporary role credentials from its metadata endpoint. Off-platform API calls then reached other memories and agents. The distinctive persistence path wrote short-term conversation events that configured extraction strategies could turn into long-term records; later retrieval directed the agent to a mutable command page. The reach was demonstrated within an account and region, not as a microVM escape or automatic cross-account compromise. The researchers observed major default-role restrictions on September 29; existing deployed roles still require independent inspection.
Why it relates: Fabricated conversation data becomes retained information used by future model decisions. The affected data is operational agent memory, not a demonstrated model training dataset.
Controlled broad-role AgentCore deployment with separate direct API and retained-memory/C2 branches; account/region scope, tool prerequisites and later default-role restrictions remain explicit.
Many recent large-scale datasets are distributed as a list of URLs pointing to individual datapoints. The researchers show that many of these datasets are vulnerable to a "split-view" poisoning attack. The attack exploits the fact that the data viewed when it was initially collected may differ from the data viewed by a user during training. The researchers identify expired and buyable domains that once hosted dataset content, making it possible to replace portions of the dataset with poisoned data. They demonstrate that for 10 popular web-scale datasets, enough of the domains are purchasable to successfully carry out a poisoning attack.
Why it relates: Published dataset URL lists identify exactly which external image origins will be downloaded. Buying expired listed domains can replace those examples, crossing the original curator-to-later-downloader integrity boundary. Public downloads were not poisoned in the study; downstream model effects were tested by local image replacement and CLIP retraining.
Many recent large-scale datasets are distributed as a list of URLs pointing to individual datapoints. The researchers show that many of these datasets are vulnerable to a "split-view" poisoning attack. The attack exploits the fact that the data viewed when it was initially collected may differ from the data viewed by a user during training. The researchers identify expired and buyable domains that once hosted dataset content, making it possible to replace portions of the dataset with poisoned data. They demonstrate that for 10 popular web-scale datasets, enough of the domains are purchasable to successfully carry out a poisoning attack.