Summary

  • This week’s news reiteration of the artificial general intelligence (AGI) possibility saw China’s DeepSeek launch a new AI model described by headed product at Hugging Face as “the most impressive AI tool I’ve ever tried”
  • New York Times columnist Ezra Klein discussed AGI on his podcast, covering its national security and labour market implications, while AGI critic Gary Marcus wrote a rebuttal arguing that AGI’s impact is over-hyped and the technology’s development is still several years away
  • Google, Scale AI and the Center for AI Safety published a paper arguing that AGI’s arrival will bring with it levels of risk not seen since the advent of atomic weapons, while Chinese researchers published their own paper proposing a survival game to assess an AGI’s ability to adapt and learn

By James O’Donnell

Original Article