Computer Security, Privacy, and DNA Sequencing...
summary
Demonstrates that synthesized DNA can encode arbitrary remote code execution by ordering strands from IDT, sequencing them on Illumina, and feeding the resulting FASTQ into a deliberately vulnerable compression utility to trigger RCE; a working exploit is ~160 bases long and arrives as real reads. Experiments show 0.1% sample bleeding (exploit reads appear in other samples 37 times) and audits find three real buffer overflows in common bioinformatics tools.
AI-written summary · about
chapters (30)
- 0:00Introduction
- 0:54Outline
- 1:23DNA Sequencing
- 3:35NextGeneration Sequencing
- 4:22How DNA Sequencing Works
- 5:02DNA Preparation
- 5:20DNA Reads
- 5:59DNA Processing
- 6:23Making Malicious DNA
- 7:34Computer Code in DNA
- 8:16The Problem
- 8:38Challenges
- 8:50Short reads
- 9:39Sequencing direction
- 10:54Sequencing noise
- 11:42Summary
- 12:20DNA Synthesis
- 13:13Repeated Sequences
- 13:44GC Content
- 14:12Secondary Structures
- 14:36Open Problems
- 15:07Final Exploit
- 15:38Conclusion
- 16:12Methodology
- 17:46Assumptions
- 18:27Side channel
- 18:38Sample multiplexing
- 19:48Data leakage
- 20:16Sequencing Concurrently
- 26:17Questions
related talks
55mHow hacking works
01h 06mHow hacking works
51mSteering Clear of Danger: Decrypting the Realities of Remote Car Hacking
47mHow to Hack a Turned-Off Computer, or Running Unsigned Code in Intel Management Engine
01h 01mSomething Something Cyber
51mExploiting Hardware Glitches And Side Channels In Perfect Software
32mRemote DNA Sequencing and Analysis; Nanopore sequencing, FPGA boards and a fortunate Epiphany
51mThe Anatomy of an Exploit