AI Beyond Identity

Did hidden image instructions bypass VLM guardrails?

First post and replies  | Last post by Mr NHI, 3 weeks ago

Did agent skills hide malicious payloads inside trusted files?

First post and replies  | Last post by Mr NHI, 3 weeks ago

What would it take for retrieval poisoning to stay contained?

First post and replies  | Last post by Mr NHI, 3 weeks ago

Could detection still stop autonomous AI agents in runtime security?

First post and replies  | Last post by Mr NHI, 3 weeks ago

Does data poisoning still slip past model drift checks?

First post and replies  | Last post by Mr NHI, 3 weeks ago

What would it take for 0.1% data poisoning to derail models?

First post and replies  | Last post by Mr NHI, 3 weeks ago

Did the Fifth Circuit footnote overstate Judge Rodriguez's AI use?

First post and replies  | Last post by Mr NHI, 3 weeks ago

What would it take to keep Fable 5 defenses usable?

First post and replies  | Last post by Mr NHI, 2 weeks ago

Does SBOM-only governance miss AI-native risk beyond code?

First post and replies  | Last post by Mr NHI, 2 weeks ago

Why is Anthropic limiting frontier capability for research?

First post and replies  | Last post by Mr NHI, 2 weeks ago