InkJect is a vulnerability tracked by ThreatCluster, appearing in 1 threat cluster built from 3 intelligence report mentions.
InkJect is a vulnerability tracked across 1 threat cluster and 3 intelligence report mentions on ThreatCluster. First observed July 1, 2026; most recent activity July 1, 2026.
DeepKeep has identified a new visual prompt injection vulnerability named 'InkJect' that affects leading visual language models (VLMs) like OpenAI's GPT-5.2 and Anthropic's Claude Sonnet 4.6. This vulnerability allows…
InkJect is a vulnerability tracked by ThreatCluster, appearing in 1 threat cluster built from 3 intelligence report mentions.
The most recent intelligence report mentioning InkJect on ThreatCluster is dated July 1, 2026.
Across ThreatCluster reporting, InkJect most frequently co-occurs with Zero-day Exploit, Anthropic, DeepKeep, OpenAI, Israel, among 9 tracked related entities.
The most significant recent cluster is “DeepKeep Reveals InkJect: New Visual Prompt Injection Vulnerability in AI Models” (4 articles · Updated July 1, 2026). InkJect appears across 1 threat cluster in total, listed above with sources.
InkJect appears in 3 intelligence report mentions across 1 deduplicated threat cluster, aggregated from 17,000+ monitored sources.