Johns Hopkins Bayview responded to a shadow AI upload with a user-side deletion and a request that the tool’s operator delete the data. No account reviewed shows the operator did, and HIPAA leaves the hospital to document what it did.
Federal supervisors are asking banks whether they can track and audit AI agent decisions and how they would shut those agents down. Bankers’ own survey answers point to legacy systems, thin AI expertise and data quality as the constraints.
US bank regulators ask whether lenders can stop their AI agents Read More »
NIST, the NSA and CISA advised developers of highly capable AI agents on Oct. 16 to preserve model weights, checkpoints and rejected tool calls for investigation when an agent takes “significant uninstructed actions” that compromise its controls. The agencies’ 14-page
Eight in 10 executives in IBM’s EMEA survey say an AI incident reached their board or C-suite in the past year, and over a third of them say it was communicated informally. The report leaves open what record those escalations left, as EU logging rules for high-risk AI approach.
The Wikimedia Foundation says agents it believes OpenAI operates made unapproved edits, probed its Etherpad and sent heavy traffic to its data services. Its account adds a victim’s records to the growing record of agent incidents.
Wikimedia publishes its own account of suspected OpenAI agent activity on its wikis Read More »
Two Boston University law professors have proposed a three-part test for AI slop that turns on effort, imposition and domain degradation rather than quality. It arrives five weeks after EU and California disclosure duties took effect on the same day.
Law professors propose a three-part test for what counts as AI slop Read More »
HaystackID’s next-generation CaseBot moves beyond conversational search to deliver structured, evidence-backed case analysis. Powered by agentic retrieval and Specialized Skill Agents, the platform generates fact narratives, chronologies, deposition kits, claim analyses and record checks directly from case data, with every factual statement linked to source evidence.
Next-generation CaseBot turns case data into instant, defensible insights Read More »
Two House committees asked DoorDash for seven categories of records by August 14, covering every Chinese-developed AI model it has evaluated or used since January 2025. The request is not a subpoena, but it reaches further than a model list, and for the first time in this investigation it names the functions expected at the briefing that follows.
DoorDash AI inquiry turns model selection into a governance test Read More »
On a preliminary record, the Ninth Circuit made the user, not Perplexity, the party accessing Amazon under the CFAA and California’s CDAFA. The records of that access can sit partly with an outside operator that no litigation hold reaches by default.
Ninth Circuit vacates Amazon injunction; AI agent logs emerge as eDiscovery issue Read More »
California’s AI Transparency Act became operative Aug. 2, the same day most of the EU’s Article 50 duties began applying. The disclosure duties are live, the platform stripping ban waits until 2027, and the provenance disclosures both regimes require can be lost in routine eDiscovery processing.
California’s AI Transparency Act arrives alongside Europe’s Article 50 Read More »
In a landmark eDiscovery decision, a federal court approved an AI-assisted review workflow and declined to require extensive disclosure of validation metrics. The ruling in Schulte v. LinkedIn signals growing judicial acceptance of AI in document review while reinforcing the importance of proportionality, reasonableness, and carefully negotiated ESI protocols
“Search, Forward” with AI: Court Approves AI-Assisted Review Workflow for eDiscovery Read More »
Tara Emory and Maura Grossman argue that GenAI TAR prompts refined through iterative testing can embody attorney work product and should rarely face compelled disclosure, with validation metrics carrying the transparency load instead.
