LESSONS ON THE USE OF TECHNOLOGY IN ELECTIONS process) required granting the requested prohibition. However, for the second prong, the court determined that for the immediately upcoming election, the requested injunctive relief could not be granted because the hardships outweighed the public interest – as the state would not be able to put a new voting mechanism or upgrades in place in time for the election. Ultimately, the Court allowed the defendant to use the machines one last time in 2019 but prohibited the state from using them after that. Further, the court directed the state to address errors in voting assignments and the voter register and mandated that each precinct has at least one physical copy of the voter registration list. The state was required to provide information to voters regarding provisional ballots, and work with a cybersecurity firm to improve and modify the voting system. The court approved the move to a new electronic voting system (ballot marking devices where the voter uses a touch screen then prints a paper ballot, which is read by an optical scanner) and directed the state to develop contingency plans that include audit provisions. This case is notable for several reasons. First, the Court relied extensively on expert testimonies and a substantial body of academic and expert practitioner material regarding the technology in question, with the court finding “a mounting tide of evidence of the inadequacy and security risks” of the outdated paperless DRE system. Second, this case had implications for the contested 2020 presidential election. Had the Court not required the Secretary of State to switch to a new auditable system in 2020, it would have impacted the ability of the State of Georgia to administer credible primaries and general elections in 2020. Given the contested nature of the 2020 election and the very close margins in Georgia, the absence of strong audit rules and a paper trail (which were in place as a result of the court’s decision) could have precipitated a major political crisis. During the 2017 general elections held in Honduras, the Organization of American States Election Observation Mission (OAS EOM) reported in detail on failings in the electronic results processing system, which crashed during results tallying. 43 Although this situation was not addressed in detail by the courts, it presents an example of the type of technical and systems-based dispute that could result in EDR proceedings. The OAS’ conclusion on the election technology was that the results processing system had significant flaws and serious breaches of IT security, and was not sufficiently strong to prevent fraud, although they could not identify a specific fraudulent incident. These concerns and other factors led to their overall conclusion that they had “observed a low-quality election process. The abundance of irregularities and deficiencies is such as to preclude full certainty regarding the outcome.” 44 There were widespread allegations of fraud and mistrust in the results, and dozens of people were killed during protests over the results. The OAS report explains the technical difficulties the EOM encountered when trying to understand what had gone wrong in the results process, and why an ICT expert was needed. Forensic evidence had not been preserved, and some evidence had been altered when software was installed onto new servers. The OAS was only able to carry out a limited, non-invasive audit, meaning they could not run tools or scripts or carry out any testing process. Their report noted, however, that 464 results forms were entered into the system by unknown persons using a non-standard method, and logs were out of sync with no explanation. 45 The OAS also reported that the company responsible for providing the scanning and transcription service had remote access to the servers, outside the EMB’s control or oversight, which posed a very high security risk. Finally, the report noted that the results processing system had OAS EOM to General Elections in Honduras 2017, Report to the Permanent Council, p. 3-4 and Electoral Technology section, p.24-29 of English version. 44 OAS EOM to General Elections in Honduras 2017, Report to the Permanent Council, p. 4. 45 This finding suggests that some limited access was provided to the system for the limited audit. 43 19