In this release, we focused on giving you more control over AI-powered changelogs, making your public URLs more reliable and consistent, and being more transparent about how we handle security and privacy.
Smarter AI options for changelog generation
Pro and Team customers now benefit from a more powerful AI model for changelog generation. By default, paid plans use GPT-5.1 for higher quality summaries, while the Free tier continues to use gpt-4o-mini so everyone can still generate AI-powered changelogs.
When regenerating a changelog entry, Pro and Team users can now choose between GPT-5.1 and gpt-4o-mini directly in the regenerate modal. This lets you pick the model that best fits each situation, for example preferring GPT-5.1 for richer summaries or gpt-4o-mini when you want faster, lighter results.
We also added an AI model selector in the AI Settings modal at the repository level for Pro and Team users. You can now set a preferred AI model for each project, which is then used automatically by background jobs that generate release summaries. Free users continue to use the default gpt-4o-mini model without additional configuration.
Behind the scenes, the AI integration now uses the appropriate API for each model, and the system respects your chosen model both when generating new entries and when regenerating existing ones. This gives you more predictable, premium-quality output where it matters most.
Cleaner, more reliable URLs and new security documentation
URL slugs for changelog entries are now consistently sanitized both in the database and in the browser. Slugs are automatically converted to lowercase, spaces and underscores become hyphens, special characters are removed, and multiple hyphens are collapsed. Leading and trailing hyphens are trimmed when you finish editing, which keeps URLs tidy.
We also fixed an issue where trailing whitespace in titles could lead to problematic slugs and potential 404 errors. Slugs generated from titles now strip leading and trailing spaces before being converted, and both the create and edit forms sanitize the slug field on the fly while preserving your cursor position so editing still feels smooth.
To give you more clarity about how we handle your data, we added a dedicated Security & Privacy documentation page and surfaced it in the docs navigation with a lock icon. This page explains what data we collect, how we encrypt and protect it, how we use third-party services like GitHub, OpenAI, and Stripe, how long we retain data, and how you can exercise your privacy rights or request account deletion.
We also refined some of the wording in the new security documentation to better match our current practices and updated contact details to use hello@autochangelog.com for security and privacy requests. This makes it easier to understand our commitments and reach out if you have questions or concerns.
Finally, the pricing and home pages now highlight Advanced AI for better changelogs as a benefit of Pro and Team plans, and “RSS feed” has been updated to “RSS feed & email notifications” so the feature descriptions more accurately reflect what you receive.