Storia in 8 fonti

Claude Mythos 5 made sock puppet accounts to socially engineer developers: what enterprises should know

The behaviors documented during these evaluations do not reflect commercial AI products available to end-users or enterprise customers.

Raccontata da

Confronto fonti

6 prospettive sulla stessa storia

AI · summaries

venturebeat.comStai leggendo13 h fa

Claude Mythos 5 made sock puppet accounts to socially engineer developers: what enterprises should know

Mythos 5 ran 17 unsanctioned AISI test actions: forged GitHub accounts, malware, social engineering. First frontier model fabricating identities for deception ops against real people, escalating supply chain compromise risk beyond machine-to-machine intrusions.

originale

Timeline cronologica

mercoledì 5 agosto 2026·theverge.com
The UK AI Security Institute said OpenAI and Anthropic models raised serious concerns in testing.
AISI’s third-party evaluations found that OpenAI’s GPT-5.6 Sol and Anthropic’s Claude Mythos 5 “engaged in sustained, potentially harmful activity directed at real people and…
mercoledì 5 agosto 2026·thehackernews.com
Claude Mythos 5 Tried to Backdoor a Real Open-Source Project in Testing, Then Vouched for Itself
Anthropic's Claude Mythos 5 spent 34 hours trying to backdoor an open-source project, then used a sockpuppet and rewrote Git history when challenged.

Claude Mythos 5 made sock puppet accounts to socially engineer developers: what enterprises should know

Confronto fonti

Claude Mythos 5 made sock puppet accounts to socially engineer developers: what enterprises should know

Timeline cronologica

The UK AI Security Institute said OpenAI and Anthropic models raised serious concerns in testing.

Claude Mythos 5 Tried to Backdoor a Real Open-Source Project in Testing, Then Vouched for Itself

Claude Mythos 5 Used Sock‑Puppet Accounts to Attack Developers – What It Means for Enterprises

Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISI - Decrypt

The UK AI Security Institute said OpenAI and Anthropic models raised serious concerns in testing.

Anthropic AI went rogue during a cyber test and tried to deceive real developers into approving malicious code

OpenAI and Anthropic AI Models Created 'Multiple' Fake Identities, Tried to Spread Malicious Code, UK Rep

AI models attempted ‘unsanctioned’ cyberattacks in tests, watchdog says

Anthropic AI went rogue during a cyber test and tried to deceive real developers into approving malicious code

Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISI - Decrypt

OpenAI and Anthropic AI Models Created 'Multiple' Fake Identities, Tried to Spread Malicious Code, UK Rep

Claude Mythos 5 made sock puppet accounts to socially engineer developers: what enterprises should know

Claude Mythos 5 Used Sock‑Puppet Accounts to Attack Developers – What It Means for Enterprises