OpenAI is rolling out a suite of election integrity measures for 2026, building on lessons from 2024’s global elections. The playbook has four pillars: directing people to reliable voting info, bolstering cyber defenses, flagging AI-generated content, and keeping ChatGPT politically neutral. Starting this fall, US and Brazil users will see live election results from the Associated Press directly in ChatGPT, while American voters get personalized voting logistics via Democracy Works. On the security front, OpenAI’s new Daybreak program gives verified voting system manufacturers access to Codex Security (which automatically patches code vulnerabilities) and Trusted Access for Cyber (frontier AI for defense). They’re also engaging with state election officials across the US. Why it matters: This marks a shift from reactive content moderation to proactive infrastructure hardening — treating election integrity as both a product design and cybersecurity challenge.