Tag
#mac
One of the largest data breaches in history was apparently twice as impactful as previously thought, with PII belonging to hundreds of millions of people sitting in the hands of cybercriminals.
### Description The vllm/model_executor/weight_utils.py implements hf_model_weights_iterator to load the model checkpoint, which is downloaded from huggingface. It use torch.load function and weights_only parameter is default value False. There is a security warning on https://pytorch.org/docs/stable/generated/torch.load.html, when torch.load load a malicious pickle data it will execute arbitrary code during unpickling. ### Impact This vulnerability can be exploited to execute arbitrary codes and OS commands in the victim machine who fetch the pretrained repo remotely. Note that most models now use the safetensors format, which is not vulnerable to this issue. ### References * https://pytorch.org/docs/stable/generated/torch.load.html * Fix: https://github.com/vllm-project/vllm/pull/12366
A critical vulnerability in Brave Browser allows malicious websites to appear as trusted sources during file uploads/downloads. Learn…
Beware of a convincing Royal Mail SMS phishing scam asking for personal details and payment for re-delivery. Learn…
“Yahoo Boy” scammers are impersonating CNN and other news organizations to create videos that pressure victims into making blackmail payments.
A previously unknown threat actor has been observed copying the tradecraft associated with the Kremlin-aligned Gamaredon hacking group in its cyber attacks targeting Russian-speaking entities. The campaign has been attributed to a threat cluster dubbed GamaCopy, which is assessed to share overlaps with another hacking group named Core Werewolf, also tracked as Awaken Likho and PseudoGamaredon.
This week on the Lock and Code podcast, host David Ruiz shares three privacy rules for 2025, and they're all about taking back control.
The MITRE framework's applied exercise provides defenders with critical feedback about how to detect and defend against common, but sophisticated, attacks.
### Summary If an attacker can control the input to the `asteval` library, they can bypass asteval's restrictions and execute arbitrary Python code in the context of the application using the library. ### Details The vulnerability is rooted in how `asteval` performs handling of `FormattedValue` AST nodes. In particular, the [`on_formattedvalue`](https://github.com/lmfit/asteval/blob/cfb57f0beebe0dc0520a1fbabc35e66060c7ea71/asteval/asteval.py#L507) value uses the [dangerous format method of the str class](https://lucumr.pocoo.org/2016/12/29/careful-with-str-format/), as shown in the vulnerable code snippet below: ```py def on_formattedvalue(self, node): # ('value', 'conversion', 'format_spec') "formatting used in f-strings" val = self.run(node.value) fstring_converters = {115: str, 114: repr, 97: ascii} if node.conversion in fstring_converters: val = fstring_converters[node.conversion](val) fmt = '{__fstring__}' if node.f...
US prosecutors charged five, including North Koreans, for tricking firms into hiring fake IT workers, sending $866K+ to…