"In preparing to release Claude Mythos, we want to act with extra caution and understand the risks it poses - even beyond what we learn in our own testing."
"The rationale seems straightforward: if today's models can already identify and even help exploit software vulnerabilities, a more capable system like Mythos could significantly accelerate both discovery and misuse."
Anthropic's advanced LLM, Mythos, was unintentionally revealed through a data leak, exposing its enhanced reasoning and coding capabilities. The leak occurred when staff mistakenly exposed information via a public data repository. Following the incident, Anthropic restricted access and confirmed the model's existence. The company aims to release Mythos carefully, focusing on potential cybersecurity risks. Initial deployment will target enterprise security teams, as the model's capabilities could significantly impact both vulnerability discovery and exploitation.
Read at Computerworld
Unable to calculate read time
Collection
[
|
...
]