The Brennan Center for Justice at NYU has spotlighted the increasing influence of artificial intelligence on election information ahead of the 2026 U.S. midterm elections. In a recent expert brief, the Center analyzed how AI chatbots impact the dissemination and perception of election-related data, emphasizing both the current limitations and potential of these technologies in combating misinformation amid changes in federal election security support.
What Happened
In a report published last week, the Brennan Center’s Elections and Government program assessed major AI chatbots such as ChatGPT, Claude, and Perplexity regarding their responses to election conspiracy theories circulating in the information ecosystem. The analysis involved “pressure testing” these AI systems by repeatedly challenging the chatbots with prominent falsehoods about election integrity. The result showed that all tested chatbots consistently disputed prevalent conspiracy theories, such as claims of widespread voter fraud, manipulated voting machines, and irregular mail ballots.
Concurrently, testimony from election security experts—including Larry Norden, Vice President of Elections and Government at the Brennan Center—highlighted that federal support structures for election security, particularly provided by the Cybersecurity and Infrastructure Security Agency (CISA), have been significantly reduced over the past 18 months. Despite this, state and local election officials have demonstrated substantial efforts to compensate for this gap ahead of the midterms.
Key Facts
The U.S. Constitution mandates that states and localities administer elections, with Congress playing a secondary role in setting minimum standards for federal elections. Following the 2016 election and documented foreign interference attempts, there was a substantial federal investment in election security infrastructure, including the creation of a collaborative information-sharing environment led by CISA.
However, federal support in the form of training, threat information sharing, and cybersecurity services has been curtailed recently, partly due to reduced trust stemming from false election claims propagated by federal officials and administrative turnover in key agencies.
Regarding AI influence, around half of Americans now reportedly use AI chatbots, with approximately 25% engaging daily. The Brennan Center’s study reveals these systems currently provide pushback against election misinformation, although AI’s sycophantic tendencies may affect their responses.
What This Means
The Brennan Center’s findings suggest that AI chatbots currently offer a net positive effect in countering widely spread election conspiracy theories, a hopeful sign in the ongoing struggle against misinformation. Their consistent rejection of false claims could provide voters with more reliable information sources than traditional social media networks, which often amplify misleading content.
Yet, the erosion of federal election security infrastructure presents challenges that AI alone cannot address. The withdrawal of CISA’s hands-on support and the breakdown of trusted relationships between federal and state officials could undermine coordinated responses to cyber threats and misinformation campaigns. This places significant responsibility on state and local governments, which, despite resource constraints, are stepping up their election security efforts.
Overall, this landscape underscores the complex interplay between technology, governance, and public trust in democratic processes. While AI offers a novel tool in combating misinformation, the broader institutional framework supporting election security remains a critical component that needs reinforcement to ensure election integrity in 2026 and beyond.
Background
Post-2016, U.S. election security became a national priority following Russian interference in the presidential election. In response, investments were made in upgrading voting infrastructure, such as introducing paper ballot backups nationwide, saving Louisiana as a notable exception. The establishment of CISA and inter-agency information sharing enabled a cooperative effort involving the private sector and national security entities to defend election systems against threats.
This federal infrastructure supported training and real-time threat intelligence sharing for election officials, fostering a relatively secure election environment up until recent reductions in support.
What Remains Unclear
The sustainability of AI chatbots’ capacity to resist manipulation by bad actors remains uncertain as adversaries may adapt and refine misinformation tactics. Additionally, the trajectory of federal election security funding and cooperation in the coming months is not yet defined, with some policymakers indicating possible reinvestment, though skepticism among state officials persists due to current mistrust.
What Comes Next
The 2026 midterm elections will provide an important testbed for both AI’s role in shaping election discourse and the adaptability of election administrations nationwide amid a shifting federal landscape. Ongoing monitoring of AI chatbot capabilities and potential regulatory or supportive actions concerning election security infrastructure will shape the lead-up to these elections.
Sources
This article is based on reporting and publicly available information from the following source:
Read more AI Regulation stories on Goka World News.