Hacker group ShinyHunters claims to have breached the Claude Mythos model, leaking screenshots of the dashboard and user data.

📄Full Article· Automatically extracted by trafilaturaGemini 翻譯1108 words

The hacker group ShinyHunters claims to have breached the internal systems of Anthropic’s Mythos model, leaking screenshots of dashboards and user data. (Previous coverage: Bloomberg reports Claude Mythos has been accessed without authorization! The hardest vulnerability for Anthropic to defend is always "people.") (Background: AquaSec: Hacker group TeamTNT launches kangaroo attacks "attempting to crack Bitcoin.") The hacker group ShinyHunters claims to have breached internal systems related to Anthropic and its Mythos model, leaking multiple screenshots online as evidence. Several KOLs on X and SlowMist Chief Security Officer 23pds have shared warnings, drawing significant attention from the cybersecurity community. The leaked screenshots cover three types of screens: user management panels, AI experiment dashboards, and model performance and cost analytics pages. However, the authenticity of these screenshots cannot be independently verified at this time. 🚨 BREAKING ShinyHunters claims to have accessed internal systems linked to Anthropic’s Mythos model — sharing screenshots of: User management panels AI experiment dashboards Model performance & cost analytics⚠️ At this stage, authenticity is unverified — but if confirmed,… pic.twitter.com/lhFEkVoQxn — Dark Web Intelligence (@DailyDarkWeb) April 22, 2026 Bloomberg claims Claude Mythos has been accessed without authorization. The timing is highly sensitive. As BlockTempo reported yesterday, Bloomberg learned from sources that on the very day Anthropic announced the launch of the Mythos Preview, an unidentified group was already using it. According to the report, the group gained access through two different channels simultaneously. The first path was human. One member of the group was employed by a third-party contractor providing services to Anthropic. This position granted them credentials to access the Mythos environment, which they shared with others in the group. The second path was the URL. Anthropic uses a fixed naming format when deploying other models; based on their understanding of this format, the group directly guessed the access location for Mythos Preview, requiring no "hacking techniques" in the traditional sense. Anthropic responded: "We are investigating a report alleging unauthorized access to the Claude Mythos Preview via a third-party vendor environment." They also stated that they have not yet found any impact on their systems from these unauthorized activities.

Data Status✓ Full text extractedRead Original (動區 BlockTempo)

🔍Historical Similar Events· Keyword + Asset Matching6 items

2026-04-23

OpenAI CEO Sam Altman slams Anthropic: Fear-mongering Claude Mythos is just to monopolize AI

Similarity 130%關鍵字 claude/mythos

2026-04-23

Anthropic Using ‘Fear-Based Marketing’ to Promote Claude Mythos: Sam Altman

Similarity 130%關鍵字 claude/mythos

2026-04-22

Claude Mythos Identifies 271 Vulnerabilities in Mozilla’s Firefox

Similarity 130%關鍵字 claude/mythos

2026-04-22