As part of pre-release safety testing for its new #GPT4 #AI model, launched Tuesday, #OpenAI allowed an AI testing group to assess the potential #risks of the model's #emergent #capabilities—including #PowerSeeking #behavior #SelfReplication, and #SelfImprovement. While the testing group found that GPT-4 was "ineffective at the autonomous replication task," the nature of the experiments raises eye-opening questions about the safety of #future #AISystems.https://arstechnica.com/information-technology/2023/03/openai-checked-to-see-whether-gpt-4-could-take-over-the-world/ #AI #Risk #safety
#gpt4 #ai #openai #risks #emergent #capabilities #powerseeking #behavior #selfreplication #selfimprovement #future #aisystems #risk #safety