June 29, 2022

ES&S Uses Undergraduate Project to Lobby New York Legislature on Risky Voting Machines

The New York State Legislature is considering a bill that would ban all-in-one voting machines. That is, voting machines that can both print votes on a ballot and scan and count votes from a ballot – all in the same paper path.

This is an important safeguard because such machines, if they are hacked by the installation of fraudulent software, can change or add votes that the voter did not intend and never got a chance to see on paper.

One voting machine company, Elections Systems and Software (ES&S), which makes an all-in-one voting machine, the ExpressVote XL, is lobbying hard against this bill. As part of its lobbying package, ES&S is claiming that “Rochester Institute of Technology researchers found zero attacks” on the ExpressVote XL, based on an article (included in ES&S’s lobbying package) from Rochester Institute of Technology entitled “RIT cybersecurity student researchers put voting machine security to the test.

If this were actually a scientific article, one could critique it as actual science.  But it’s not a scientific paper:  The article is written by Scott Bureau, Senior Communications Specialist, RIT Marketing and Communications in the RIT public relations department. 

The article describes an undergraduate student “capstone project.”  The students were interviewed by ES&S, allowed ES&S to inspect their testing site, and then signed a nondisclosure agreement with ES&S.  The students made up two attack scenarios, then spent 10 days trying to find attacks.  They found some vulnerabilities, but not one that could change votes.

The students made public a one-page poster describing their project. It’s fine for undergraduate student work; capstone projects are a really useful part of engineering education.  But it’s not a scientific paper that describes their methods, the limitations placed upon them by needing permission from ES&S, or, in any detail – their results.

Even so, the students describe enough for me to notice that they missed three of the most important attack scenarios:

  • Hacker intrusion into the ES&S corporate engineering network, stealing cryptographic keys and source code, or altering the software to be installed into all ExpressVote XL machines nationwide in the next software update.
  • Hacker intrusion into the county election administrator’s network, stealing cryptographic keys and allowing manipulation of ballot-definition downloads.
  • Stealing an ExpressVote XL anywhere in the country, not just in New York, and tearing it apart to reverse engineer and steal crypto keys.
  • There may be many other attacks.  That’s why penetration testing can never prove that a computer system is secure: pen-testing only examines the attacks that the pen-testers happen to think of.

These are standard attacks. These are the ones that can be so effective and dangerous that there is good reason for banning such voting machines.    Maybe those Rochester students are aware of such attacks. Maybe not. But it seems unlikely that ES&S would have given permission for such experiments. That’s why respectable academic security researchers don’t restrict their activities to those in the comfort zone of the corporations whose products they are examining.It is irresponsible and misleading of ES&S to characterize an undergraduate student project, conducted under conditions controlled by ES&S, described in a publicity puff-piece written by a public-relations flack, as “RIT researchers found zero attacks.”

Will Web3 Follow in the Footsteps of the AI Hype Cycle?

For many, the global financial crisis of 2008 marked a turning point for trust in established institutions. It is unsurprising that during this same historical time period, Bitcoin, a decentralized cryptocurrency that aspired to operate independent from state manipulation, began gaining traction. Since the birth of Bitcoin, other decentralized technologies have been introduced that enable a broader range of functionalities including decentralized finance (DeFi), non-fungible tokens (NFTs), a wide range of other cryptocurrencies, and decentralized autonomous organizations (DAOs). 

These types of technologies constitute what is sometimes referred to as “web3.” In contrast to web2, our current version of the web, which relies heavily on centralized platforms and corporate intermediaries–think Facebook’s social network or Amazon’s webshop–web3 promises to redistribute power and agency back into the hands of users through decentralized peer-to-peer technology. Although web3 has garnered fervent support and equally fervent critique, it is undeniable that cryptocurrencies and other decentralized technologies have captured the mainstream imagination. 

What is less clear is whether the goals and practices of emerging businesses in the web3 sector align with, or stand in conflict with, the ideologies of web3’s most enthusiastic supporters. Organizational sociology has long established that organizations’ external rhetoric, which is shaped by a field’s perception of what is culturally and socially legitimate, may not fully align with their internal rhetoric or day-to-day practices. Continuing in this tradition, in a recent study, my colleague at Princeton’s Center for Information Technology Policy, researcher Elizabeth Watkins, and I sought to understand how people working at artificial intelligence (AI) startups think about, build, and publicly discuss their technology. We conducted interviews with 23 individuals working at early-stage AI startups across a variety of industry domains including healthcare, agriculture, business intelligence, and others. We asked them about how their AI works as well as about the pressures they face as they try to grow their companies.

In our interviews, the most prevalent theme we observed was that startup founders and employees felt they needed to hype up their AI to potential investors and clients. Widespread narratives about the transformative potential of AI have led non-AI savvy stakeholders to have unrealistic expectations about what AI can do– expectations that AI startups must contend with to gain market adoption. Some, for instance, have resorted to presenting artificially inflated estimates of their models’ performance to satisfy the demands of investors or clients that don’t really understand how models work or how they should be evaluated. From the perspective of the startup entrepreneurs we interviewed, if other AI startups promise the moon, it is difficult for their companies to compete if all they promise is a moon-shaped rock, especially if potential clients and investors cannot tell the difference. At the same time, these startup entrepreneurs did not actually buy into the hype themselves. Afterall, as AI practitioners, they know as well as any other tech skeptic what the limitations of AI are. 

In our AI startups study, several participants likened the hype surrounding AI to the hype that also surrounds blockchain, the backbone that undergirds decentralized technology. Yet unlike AI companies who hope to disrupt existing modes of performing tasks, hardline web3 evangelists see decentralized technology as a mechanism for disrupting the existing social, political, and economic order. That kind of disruption would take place on an entirely different scale than AI companies attempting to make tedious or boring tasks a little more automatic. But are web3 businesses actually hoping to effect the same kind of wide sweeping societal change web3 evangelists are hoping for?

In a study I’m kicking off with Johannes Lenhard, an anthropologist at the University of Cambridge who studies venture capital investors, we aim to understand where the ideological rubber of web3 meets the often unforgiving road to commercial success. We will interview entrepreneurs working at web3 businesses and investors working at investment firms with a focus on web3. Through these interviews, we aim to understand what their ideological visions of web3 are and the extent to which they have been able to realize those visions into real-world technology and business practices. 

As a preliminary glimpse into these questions, I did a quick and dirty analysis* of content from the blogs that Andreessen Horowitz (a16z), a prominent venture capital firm, posted about the companies in their web3 portfolio (top image). In order to get insight into the rhetoric of the companies themselves, I also looked at content from the landing pages of several of a16z’s web3 portfolio companies (bottom image). Visualization of the most frequently used terms of both data sources are below where bigger words are those that are used more frequently.

Word cloud from a16z’s blog posts

Word cloud from portfolio companies’ landing pages

Although this analysis is by no means scientific, it suggests that whereas companies’ external rhetoric emphasizes technical components, investors’ external rhetoric emphasizes vision. 

We don’t yet know whether we will observe these kinds of trends in our new study, but we hope to gain deeper empirical insights into both the public facing discourse of web3 stakeholder groups as well as into the rhetoric they use internally to shape their own self-perception and practices. Will blockchain shepherd in a newer, more democratic version of the web? A borderless society? Decentralized governance by algorithms? Or will it instead deliver only a few interesting widgets and business as usual? We’ll report back when we find out!

Interested in hearing more about the study or participating? Send me an email at .

*analysis performed on March 9th, 2022

 A Multi-pronged Strategy for Securing Internet Routing

By Henry Birge-Lee, Nick Feamster, Mihir Kshirsagar, Prateek Mittal, Jennifer Rexford

The Federal Communications Commission (FCC) is conducting an inquiry into how it can help protect against security vulnerabilities in the internet routing infrastructure. A number of large communication companies have weighed in on the approach the FCC should take. 

CITP’s Tech Policy Clinic convened a group of experts in information security, networking, and internet policy to submit an initial comment offering a public interest perspective to the FCC. This post summarizes our recommendations on why the government should take a multi-pronged strategy to promote security that involves incentives and mandates. Reply comments from the public are due May 11.

The core challenge in securing the internet routing infrastructure is that the original design of the network did not prioritize security against adversarial attacks. Instead, the original design focused on how to route traffic through decentralized networks with the goal of delivering information packets efficiently, while not dropping traffic. 

At the heart of this routing system is the Border Gateway Protocol (BGP), which allows independently-administered networks (Autonomous Systems or ASes) to announce reachability to IP address blocks (called prefixes) to neighboring networks. But BGP has no built-in mechanism to distinguish legitimate routes from bogus routes. Bogus routing information can redirect internet traffic to a strategic adversary, who can launch a variety of attacks, or the bogus routing can lead to accidental outages or performance issues. Network operators and researchers have been actively developing measures to counteract this problem.

At a high level, the current suite of BGP security measures depend on building systems to validate routes. But for these technologies to work, most participants have to adopt them or the security improvements will not be realized. In other words, it has many of the hallmarks of a “chicken and egg” situation. As a result, there is no silver bullet to address routing security.

Instead, we argue, the government needs a cross-layer strategy that embraces pushing different elements of the infrastructure to adopt security measures that protect legitimate traffic flows using a carrot-and-stick approach. Our comment identifies specific actions Internet Service Providers, Content Delivery Networks and Cloud Providers, Internet Exchange Points, Certificate Authorities, Equipment Manufacturers, and DNS Providers should take to improve security. We also recommend that the government funds and supports academic research centers that collect real-time data from a variety of sources that measure traffic and how it is routed across the internet.  

We anticipate several hurdles to our recommended cross-layer approach: 

First, to mandate the cross-layer security measures, the FCC has to have regulatory authority over the relevant players. And, to the extent a participant does not fall under the FCC’s authority, the FCC should develop a whole-of-government approach to secure the routing infrastructure.

Second, large portions of the internet routing infrastructure lie outside the jurisdiction of the United States. As such, there are international coordination issues that the FCC will have to navigate to achieve the security properties needed. That said, if there is a sufficient critical mass of providers who participate in the security measures, that could create a tipping point for a larger global adoption.

Third, the package of incentives and mandates that the FCC develops has to account for the risk that there will be recalcitrant small and medium sized firms who might undermine the comprehensive approach that is necessary to truly secure the infrastructure.

Fourth, while it is important to develop authenticated routes for traffic to counteract adversaries, there is an under-appreciated risk from a flipped threat model – the risk that an adversary takes control of an authenticated node and uses that privileged position to disrupt routing. There are no easy fixes to this threat – but an awareness of this risk can allow for developing systems to detect such actions, especially in international contexts.