4 Ways Generative AI is Strengthening Democracies Worldwide
- Synerf
- Nov 23
- 4 min read
Source Note: This analysis is based on original reporting by Bruce Schneier for theguardian.com
TL;DR
AI is a Power Distributor, not just a Concentrator: The most effective uses of AI in democracy are to amplify citizen and civil servant power, not replace humans.
Focus on Transparency & Engagement: Examples show AI facilitating better voter information, increased litigation as civic action, and enhanced governmental oversight.
Key GE Tools to Know: Mirai Assembly (Japan), AI Tip Sheets (US), and the Apertus public AI model (Switzerland).
The Future is "Public AI": Open-source, democratically controlled AI models, like Apertus, are essential to mitigate the risks of proprietary "Big Tech" AI systems in governance.
Generative AI is colliding with democracy. While the prevailing narrative focuses on the risks, from undermining the information ecosystem to enabling authoritarian consolidation, there are also significant opportunities.
The core principle behind successful democratic AI use is power distribution. The technology is being used to assist people performing their democratic tasks, not to replace them. This includes politics, litigation, voting, and watchdog journalism.
Here are real-world examples from around the world showing how AI is being used to make democracy better, stronger, and more responsive.
Case Study 1: Japan — AI Avatars for Political Engagement
How is Japan Leveraging AI for Citizen Outreach?
In Japan, AI is being used to create new, highly accessible channels for political engagement.
In 2024, engineer Takahiro Anno ran as an independent candidate for governor of Tokyo. His campaign was distinguished by the unprecedented use of an authorized AI avatar.
This avatar answered 8,600 questions from voters on a continuous 17-day YouTube livestream.
This level of direct, non-stop engagement garnered attention from campaign innovators globally.
The Mirai Assembly App
Anno-san was recently elected to Japan’s upper legislative chamber, again leveraging AI to answer over 20,000 questions from constituents.
His new party, Team Mirai, is an AI-enabled civic technology shop.
They are developing the Mirai Assembly app .
Function: This app allows constituents to express opinions and ask questions about bills in the legislature.
Impact: The AI organizes these public expressions, and the party promises its members will direct their questioning in committee hearings based on this public input.
This is a clear example of AI amplifying the power of the constituent voice.
Case Study 2: Brazil — AI for Judicial Efficiency and Civic Action
How is Brazil Using AI in its Court System?
Brazil is a notoriously litigious country, which has resulted in chronically overwhelmed courts. Estimates show the Brazilian federal government spends about 1.6% of GDP per year operating the courts, plus another 2.5% to 3% of GDP on court-ordered payments.
Since 2019, the Brazilian government has aggressively adopted AI to automate procedures throughout its judiciary.
AI’s Role in the Judiciary:
The AI is not making judicial decisions. It is performing supporting, efficiency-driven tasks:
Distributing caseloads.
Performing legal research.
Transcribing hearings.
Identifying duplicative filings.
Preparing initial orders for signature.
Clustering similar cases for joint consideration.
Key Factual Result:
The results have been significant. Brazil’s federal supreme court backlog dropped in 2025 to its lowest levels in 33 years.
Case Study 3: United States — AI for Watchdog Journalism
How is AI Supporting Investigative Journalism in the US?
In California, the non-profit, nonpartisan news organization CalMatters runs the Digital Democracy project.
Digital Democracy collects all public utterances of California elected officials: floor speeches, committee comments, social media posts, voting records, legislation, and campaign contributions. All of this information is available on a free online platform.
The AI Tip Sheets Feature
In 2024, CalMatters launched a new AI-powered feature: AI Tip Sheets .
Function: The AI searches through all the collected political data looking for anomalies.
Examples of Anomalies: A sudden change in a voting position linked to a large campaign contribution.
Impact: These anomalies appear on a webpage for journalists, providing them with story ideas and evidence-based data for further reporting.
This is a critical innovation. It is not AI replacing human journalists, but a civic watchdog organization using technology to feed evidence-based insights to human reporters, helping a reduced number of journalists maintain the scope of action our democracy relies on.
The Emergence of "Public AI" and Democratic Alternatives
A significant barrier to the broader adoption of AI in democratic processes is the AI market itself. Core technologies are largely developed by US tech giants. These are proprietary, black-box systems whose training data, guardrails, and inherent biases are unknown and not subject to democratic control. This is an unacceptable risk for use in sensitive democratic contexts.
The Solution: Public AI
Advocates for the development of "public AI" support models and systems that are developed under democratic control and deployed for public benefit, not for corporate shareholders.
Apertus: The Swiss Model for Public AI
Switzerland has recently released the world's most powerful and fully realized public AI model, called Apertus.
Development: Developed jointly by the Swiss government and the university ETH Zurich.
Open-Source: It is entirely open source : open data, open code, open weights, and free for anyone to use.
Ethical Training: No illegally acquired copyrighted works were used in its training, and it did not exploit poorly paid human labor.
Proof of Concept: Apertus demonstrates that achieving high performance does not require spending trillions of dollars, offering a democratic alternative to Big Tech-controlled corporate AI.
Comments