#disassemblers — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #disassemblers, aggregated by home.social.
-
@MishaVelthuis : firstly, I am definitely not an AI expert.
W.r.t. source code analysis: in fact one does not need "source code" (e.g. higher level languages) to be able to spot bugs. More or less like you wrote, in the end every computer program contains sequences of instructions for the targeted CPU (including calls to, possibly documented, operating system and other libraries).
Reverse engineers love decompilers because assembler code is a lot harder to interpret *by humans* than higher level source code (it can be done, I did it approx. 30 - 40 years ago for various CPU architectures, but it is very time consuming and error-prone).
Malware typically consist of (highly) obfuscated code to fool decompilers and emulators. I've not studied it but probably the anti-malware industry is already using AI to help them analyse malware (and, feeding the rat race, quite likely malware makers are using AI to obfuscate their code).
Internet search terms such as "malware analysis llm" may possibly help you to find related articles.
Regardless, IMO not possessing the original source code should not pose a major hurdle for finding bugs in executable code.
#AI #ReverseEngineering #Malware #DeCompilers #IDA #DisAssemblers #Infosec #AntiMalware
-
@MishaVelthuis : firstly, I am definitely not an AI expert.
W.r.t. source code analysis: in fact one does not need "source code" (e.g. higher level languages) to be able to spot bugs. More or less like you wrote, in the end every computer program contains sequences of instructions for the targeted CPU (including calls to, possibly documented, operating system and other libraries).
Reverse engineers love decompilers because assembler code is a lot harder to interpret *by humans* than higher level source code (it can be done, I did it approx. 30 - 40 years ago for various CPU architectures, but it is very time consuming and error-prone).
Malware typically consist of (highly) obfuscated code to fool decompilers and emulators. I've not studied it but probably the anti-malware industry is already using AI to help them analyse malware (and, feeding the rat race, quite likely malware makers are using AI to obfuscate their code).
Internet search terms such as "malware analysis llm" may possibly help you to find related articles.
Regardless, IMO not possessing the original source code should not pose a major hurdle for finding bugs in executable code.
#AI #ReverseEngineering #Malware #DeCompilers #IDA #DisAssemblers #Infosec #AntiMalware
-
@MishaVelthuis : firstly, I am definitely not an AI expert.
W.r.t. source code analysis: in fact one does not need "source code" (e.g. higher level languages) to be able to spot bugs. More or less like you wrote, in the end every computer program contains sequences of instructions for the targeted CPU (including calls to, possibly documented, operating system and other libraries).
Reverse engineers love decompilers because assembler code is a lot harder to interpret *by humans* than higher level source code (it can be done, I did it approx. 30 - 40 years ago for various CPU architectures, but it is very time consuming and error-prone).
Malware typically consist of (highly) obfuscated code to fool decompilers and emulators. I've not studied it but probably the anti-malware industry is already using AI to help them analyse malware (and, feeding the rat race, quite likely malware makers are using AI to obfuscate their code).
Internet search terms such as "malware analysis llm" may possibly help you to find related articles.
Regardless, IMO not possessing the original source code should not pose a major hurdle for finding bugs in executable code.
#AI #ReverseEngineering #Malware #DeCompilers #IDA #DisAssemblers #Infosec #AntiMalware
-
@MishaVelthuis : firstly, I am definitely not an AI expert.
W.r.t. source code analysis: in fact one does not need "source code" (e.g. higher level languages) to be able to spot bugs. More or less like you wrote, in the end every computer program contains sequences of instructions for the targeted CPU (including calls to, possibly documented, operating system and other libraries).
Reverse engineers love decompilers because assembler code is a lot harder to interpret *by humans* than higher level source code (it can be done, I did it approx. 30 - 40 years ago for various CPU architectures, but it is very time consuming and error-prone).
Malware typically consist of (highly) obfuscated code to fool decompilers and emulators. I've not studied it but probably the anti-malware industry is already using AI to help them analyse malware (and, feeding the rat race, quite likely malware makers are using AI to obfuscate their code).
Internet search terms such as "malware analysis llm" may possibly help you to find related articles.
Regardless, IMO not possessing the original source code should not pose a major hurdle for finding bugs in executable code.
#AI #ReverseEngineering #Malware #DeCompilers #IDA #DisAssemblers #Infosec #AntiMalware
