Over the past months, I’ve been developing *AIHint*, an open standard for verifiable website trust metadata. The idea is simple: each website can publish a signed JSON “hint” that provides transparent, tamper-proof information about itself.

Why? Because today, AI systems, bots, and even humans often struggle to know whether a website is authentic, safe, or trustworthy. AIHint aims to provide a *cryptographically verifiable trust layer* that works across platforms.

Key points: - *Signed metadata*: each “hint” is digitally signed, ensuring integrity. - *Universal format*: JSON-based, easy to integrate in any ecosystem. - *Trust scoring*: cross-platform engine available in Python, PHP, and JS. - *Use cases*: AI content filtering, cybersecurity, compliance checks, or simply displaying trust indicators in apps.

This is still early, but the spec and first implementations are open source here: https://github.com/Ai-Hint/aihint-standard Docs: https://docs.aihint.org

I’d love feedback from the community: - What do you think about standardized trust scoring for websites? - Any pitfalls or design flaws we should anticipate? - Would this be useful in your projects?

Contributors and critical feedback are very welcome