Tag Archives: security

How to Spread Disinformation with Unicode

Academic papers, Security engineering, Web security, ,

There are many different ways to represent the same text in Unicode. We’ve previously exploited this encoding-visualization gap to craft imperceptible adversarial examples against text-based machine learning systems and invisible vulnerabilities in source code.

In our latest paper, we demonstrate another attack that exploits the same technique to target Google Search, Bing’s GPT-4-powered chatbot, and other text-based information retrieval systems.

Consider a snake-oil salesman trying to promote a bogus drug on social media. Sensible users would do a search on the alleged remedy before ordering it, and sites containing false information would normally be drowned out by genuine medical sources in modern search engine rankings. 

But what if our huckster uses a rare Unicode encoding to replace one character in the drug’s name on social media? If a user pastes this string into a search engine, it will throw up web pages with the same encoding. What’s more, these pages are very unlikely to appear in innocent queries.

The upshot is that an adversary who can manipulate a user into copying and pasting a string into a search engine can control the results seen by that user. They can hide such poisoned pages from regulators and others who are unaware of the magic encoding. These techniques can empower propagandists to convince victims that search engines validate their disinformation.

Talking Trojan: Analyzing an Industry-Wide Disclosure

Academic papers, Open-source security, Programming languages, Security economics, Security engineering, , ,

Talking Trojan: Analyzing an Industry-Wide Disclosure tells the story of what happened after we discovered the Trojan Source vulnerability, which broke almost all computer languages, and the Bad Characters vulnerability, which broke almost all large NLP tools. This provided a unique opportunity to measure software maintenance in action. Who patched quickly, reluctantly, or not at all? Who paid bug bounties, and who dodged liability? What parts of the disclosure ecosystem work well, which are limping along, and which are broken?

Security papers typically describe a vulnerability but say little about how it was disclosed and patched. And while disclosing one vulnerability to a single vendor can be hard enough, modern supply chains multiply the number of affected parties leading to an exponential increase in the complexity of the disclosure. One vendor will want an in-house web form, another will use an outsourced bug bounty platform, still others will prefer emails, and *nix OS maintainers will use a very particular PGP mailing list. Governments sort-of want to assist with disclosures but prefer to use yet another platform. Many open-source projects lack an embargoed disclosure process, but it is often in the interest of commercial operating system maintainers to write embargoed patches – if you can get hold of the right people.

A vulnerability that affected many different products at the same time and in similar ways gave us a unique chance to observe the finite-impulse response of this whole complex system. Our observations reveal a number of weaknesses, such as a potentially dangerous misalignment of incentives between commercially sponsored bug bounty programs and multi-vendor coordinated disclosure platforms. We suggest tangible changes that could strengthen coordinated disclosure globally.

We also hope to inspire other researchers to publish the mechanics of individual disclosures, so that we can continue to measure and improve the critical ecosystem on which we rely as our main defense against growing supply chain threats. In the meantime, our paper can be found here, and will appear in SCORED ‘22 this November.

Security course at Cambridge

Uncategorized, , , ,

I have taken over the second-year Security course at Cambridge, which is traditionally taught in Easter term. From the end of April onwards I will be teaching three lectures per week. Taking advantage of the fact that Cambridge academics own the copyright and performance rights on their lectures, I am making all my undergraduate lectures available at no charge on my YouTube channel frankstajanoexplains.com. My lecture courses on Algorithms and on Discrete Mathematics are already up and I’ll be uploading videos of the Security lectures as I produce them, ahead of the official lecturing dates. I have uploaded the opening lecture this morning. You are welcome to join the class virtually and you will receive exactly the same tuition as my Cambridge students, at no charge. 


The philosophy of the course is to lead students to learn the fundamentals of security by “studying the classics” and gaining practical hands-on security experience by recreating and replicating actual attacks. (Of course the full benefits of the course are only reaped by those who do the exercises, as opposed to just watching the videos.)


This is my small contribution to raising a new generation of cyber-defenders, alongside the parallel thread of letting young bright minds realise that security is challenging and exciting by organising CTFs (Capture-The-Flag competitions) for them to take part in, which I have been doing since 2015 and continue to do. On that note, any students (undergraduate, master or PhD) currently studying in a university in UK, Israel, USA, Japan, Australia and France still have a couple more days to sign up for our 2022 Country to Country CTF, a follow-up to the Cambridge to Cambridge CTF that I co-founded with Howie Shrobe and Lori Glover at MIT in 2015. The teams will mix people at different levels so no prior experience is required. Go for it!

CoverDrop: Securing Initial Contact for Whistleblowers

Academic papers, Privacy technology,

Whistleblowing is dangerous business. Whistleblowers face grave consequences if they’re caught and, to make matters worse, the anonymity set – the set of potential whistleblowers for a given story – is often quite small. Mass surveillance regimes around the world don’t help matters either. Yet whistleblowing has been crucial in exposing corruption, rape and other crimes in recent years. In our latest research paper, CoverDrop: Blowing the Whistle Through A News App, we set out to create a system that allows whistleblowers to securely make initial contact with news organisations. Our paper has been accepted at PETS, the Privacy Enhancing Technologies Symposium.

To work out how we could help whistleblowers release sensitive information to journalists without exposing their identity, we conducted two workshops with journalists, system administrators and software engineers at leading UK-based news organisations. These discussions made it clear that a significant weak point in the whistleblowing chain is the initial contact by the source to the journalist or news organisation. Sources would often get in touch over insecure channels (e.g., email, phone or SMS) and then switch to more secure channels (e.g., Tor and Signal) later on in the conversation – but by then it may be too late. 

Existing whistleblowing solutions such as SecureDrop rely on Tor for anonymity and expect a high degree of technical competence from its users. But in many cases, simply connecting to the Tor network is enough to single out the whistleblower from a small anonymity set. 

CoverDrop takes a different approach. Instead of connecting to Tor, we embed the whistleblowing mechanism in the mobile news app published by respective news organisations and use the traffic generated by all users of the app as cover traffic, hiding any messages from whistleblowers who use it. We implemented CoverDrop and have shown it to be secure against a global passive network adversary that also has the ability to issue warrants on all infrastructure as well as the source and recipient devices.

We instantiated CoverDrop in the form of an Android app with the expectation that news organisations embed CoverDrop in their standard news apps. Embedding CoverDrop into a news app provides the whistleblower with deniability as well as providing a secure means of contact to all users. This should nudge potential whistleblowers away from using insecure methods of initial contact. The whistleblowing component is a modified version of Signal, augmented with dummy messages to prevent traffic analysis. We use the Secure Element on mobile devices, SGX on servers and onion encryption to reduce the ability of an attacker to gain useful knowledge even if some system components are compromised.

The primary limitation of CoverDrop is its messaging bandwidth, which must be kept low to minimise the networking cost borne by the vast majority of news app users who are not whistleblowers. CoverDrop is designed to do a critical and difficult part of whistleblowing: establishing initial contact securely. Once a low-bandwidth communication channel is established, the source and the journalist can meet in person, or use other systems to send large documents.

The full paper can be found here.

Mansoor Ahmed-Rengers, Diana A. Vasile, Daniel Hugenroth, Alastair R. Beresford, and Ross Anderson. CoverDrop: Blowing the Whistle Through A News App. Proceedings on Privacy Enhancing Technologies, 2022.

CHERI: Architectural support for the scalable implementation of the principle of least privilege

Academic papers, Open-source security, Operating systems, Processors, Programmable logic, Programming languages, , , , ,

[CHERI tablet photo]
FPGA-based CHERI prototype tablet — a 64-bit RISC processor that boots CheriBSD, a CHERI-enhanced version of the FreeBSD operating system.
Only slightly overdue, this post is about our recent IEEE Security and Privacy 2015 paper, CHERI: A Hybrid Capability-System Architecture for Scalable Software Compartmentalization. We’ve previously written about how our CHERI processor blends a conventional RISC ISA and processor pipeline design with a capability-system model to provide fine-grained memory protection within virtual address spaces (ISCA 2014, ASPLOS 2015). In our this new paper, we explore how CHERI’s capability-system features can be used to implement fine-grained and scalable application compartmentalisation: many (many) sandboxes within a single UNIX process — a far more efficient and programmer-friendly target for secure software than current architectures.

Continue reading CHERI: Architectural support for the scalable implementation of the principle of least privilege →

Plaintext Password Reminders

Authentication, Web security, ,

There was a public outcry followed by ICO “making enquiries” when Troy Hunt published a post about Tesco’s plaintext password reminders exactly a month ago.

I wanted to use the reference for a text I was writing last week when someone asked me about online accounts of Companies House. At that moment I said to myself, wait a second. Companies House sends plaintext reminders as well. How strange. I sent a link to a short post to ComputerWorld. They in turn managed to get a statement from Companies House that includes:

“… although it is [Companies House] certified to the ISO 27001 standard and adheres to the government’s Security Policy Framework, it will carry out a review of its systems in order to establish whether there is a threat to companies’ confidential information.” Continue reading Plaintext Password Reminders →