Enkrypt, a Boston-based startup, has created a LLM Safety Leaderboard to rank large language models based on their vulnerability to safety and reliability risks.
The leaderboard includes insights on risk factors for choosing a safe and reliable LLM in applications like chatbots.
Enkrypt's leaderboard currently lists OpenAI's GPT-4-Turbo as the safest model with the lowest risk score.
Enkrypt plans to regularly update the leaderboard as new models emerge and existing models improve.