Liam Romanis

I am this person!

Choose people for your Council

Their For/Against majority represents you.

Unless you vote directly.

Independent cybersecurity and AI security consultant; author of a draft model-card security standard.

Speaks on: AI Ethics AI Governance AI Regulation AI Safety Open Science

I am this person!

ai-ethics (1) ai-governance (1) ai-regulation (1) ai-safety (1) open-science (1) open-source (1) policy (1) regulations (1) research-policy (1) transparency (1)

Top New

Require open source AI models to include safety benchmarks in release notes
47 opinions

Liam Romanis votes For and says:

I am this person! Add a sourced quote Report comment

This standard applies to any machine learning model published to a public model repository with a downloadable model card or README. [...] MCSS-1 (Baseline) | Any publicly published model | All HuggingFace models, open-source releases [...] Safety ev...
more AI Verified source (Apr 20, 2026)

added 5d ago

Other authors to follow

Dario Amodei CEO at Anthropic OpenAI AI research organization Anthropic AI safety research company Sam Altman CEO at OpenAI Yoshua Bengio AI Pioneer, Turing Award winner Demis Hassabis Nobel laureate, AI Researcher and CEO of DeepMind Bernie Sanders U.S. Senator from Vermont Geoffrey Hinton Godfather of Deep Learning Pope Leo XIV Head of the Catholic Church