Challenges and implications of verifiable builds for security-critical open-source software
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The majority of computer users download compiled software and run it directly on their machine. Apparently, this is also true for open-sourced software -- most users would not compile the available source, and implicitly trust that the available binaries have been compiled from the published source code (i.e., no backdoor has been inserted in the binary). To verify that the official binaries indeed correspond to the released source, one can compile the source of a given application, and then compare the locally generated binaries with the developer-provided official ones. However, such simple verification is non-trivial to achieve in practice, as modern compilers, and more generally, toolchains used in software packaging, have not been designed with verifiability in mind. Rather, the output of compilers is often dependent on parameters that can be strongly tied to the building environment. In this paper, we analyze a widely-used encryption tool, TrueCrypt, to verify its official binary with the corresponding source. We first manually replicate a close match to the official binaries of sixteen most recent versions of TrueCrypt for Windows up to v7.1a, and then explain the remaining differences that can solely be attributed to non-determinism in the build process. Our analysis provides the missing guarantee on official binaries that they are indeed backdoor-free, and makes audits on TrueCrypt's source code more meaningful. Also, we uncover several sources of non-determinism in TrueCrypt's compilation process; these findings may help create future verifiable build processes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it