The Anthropic-Singapore AI Safety Institute Technical AI Evals Workshop on 12 October will be held at Lorong AI alongside Singapore International Cyber Week, IMDA’s AI Assurance Exchange, and Anthropic’s Singapore office opening.

The workshop will bring together Anthropic, government/AISI representatives, and members of Singapore’s AI safety and research community for a technical exchange on AI evaluations.

Programme

  • 9.00 AM: Registration

  • 9.30 AM: Opening remarks by Anthropic co-founder Ben Mann, International Policy Lead Michael Sellitto, and SG AISI/IMDA senior executives

  • 10.00 AM onwards: Technical presentations by Anthropic experts- 10.00 AM: Safety evals, best practices and safeguards – Evan Frondorf, Safeguards

  • 10.45 AM: Multilingual evals – Geoff Bacon, Multilingual Evals (remote)

  • 11.30 AM: Cyber risks and evals – Andrew Fasano, Anthropic Frontier Red Team (remote)

  • 12.15 PM: CBRNE evals – Ivana Cvijovic (remote)

  • 1.00 PM: Bites and networking