From 535a9fb04949db09a99d4793b6e3c3e55900e77f Mon Sep 17 00:00:00 2001 From: jaseg Date: Fri, 21 Nov 2025 19:11:20 +0100 Subject: [PATCH] Include some of Benny's feedback --- ai-llm-use-disclosure.tex | 2 +- chapter-introduction/chapter.tex | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/ai-llm-use-disclosure.tex b/ai-llm-use-disclosure.tex index 6d2d38b..67cc6b2 100644 --- a/ai-llm-use-disclosure.tex +++ b/ai-llm-use-disclosure.tex @@ -10,7 +10,7 @@ wrong output. Thus, I used the following list of observations to guide my LLM us \begin{enumerate} \item Passing text through an LLM is an imprecise operation. Especially when large amounts of text are passed - through an LLM, despite clear instructions such as ``only fix spelling errors'', the LLM output might deviate + through an LLM, despite clear instructions such as ``only fix spelling errors,'' the LLM output might deviate from the source text. Therefore, the document text should never be passed through the LLM, and the LLM should be prompted to point out problems, or to produce a list of suggestions for improvements instead. \item LLMs are really bad at summarizing text that contains novel concepts. LLM summaries of text often converge to diff --git a/chapter-introduction/chapter.tex b/chapter-introduction/chapter.tex index bb30706..fc10acd 100644 --- a/chapter-introduction/chapter.tex +++ b/chapter-introduction/chapter.tex @@ -12,7 +12,7 @@ All Cops Are Bastards, or ACAB is a slogan popular in far left and anarchist cir that expresses a rejection of state authority~\cite{constantinouAppliedResearchPolicing2021}. While politically, this blanket rejection is a fringe viewpoint with no mainstream acceptance, there exists a parallel between this and modern cryptographic best practice. In modern cryptography, it is generally seen as best practice to have the least amount of -keys possible involved in any computation. and cryptographers have time and time again strongly rejected attempts by +keys possible involved in any computation and cryptographers have time and time again strongly rejected attempts by states and other authorities to insert backdoor access mechanisms into cryptographic systems~\cite{ abelsonRisksKeyRecovery1997, abelsonKeysDoormats2015, @@ -34,7 +34,7 @@ the system design, its possibility is considered a vulnerability. % required to comply with so-called \emph{Lawful Interception} orders on particular customers or traffic types, and % datacenter operators commonly provide hardware access to state authorities. The design decisions in cryptographic % protocols generally hold, and the gold standard for backdoor access to modern systems is either exploiting a -% \emph{zero-day} flaw that is not yet publically known, or acquiring physical access to the target system. +% \emph{zero-day} flaw that is not yet publicly known, or acquiring physical access to the target system. In this thesis, we aim to extend the level of protection afforded by cryptographic protocol design down the technology stack. While cryptographic protocols and modern software from the operating system up make it possible to secure the