According to reports, in a paper in the British weekly "New Scientist", scientists said that artificial intelligence has a 5% chance of causing human extinction.The author of the paper, Katya Glass of the California Institute of Machine Intelligence, said,Most AI researchers do not find it implausible that advanced AI could destroy humanity.I think this common perception that there is a non-trivial risk is more telling than the exact percentage risk.

Compared to answers from the same survey conducted in 2022, many AI researchers predict that AI will reach certain milestones earlier than previously predicted.

This coincides with the advent of the Chat Generation Pre-trained Transformer (ChatGPT) in November 2022 and Silicon Valley’s rush to widely deploy AI chatbot-like services based on large language models.

also,By 2047, there is a 50% chance that AI will outperform humans on every task.

By 2116, there is a 50% chance that all human jobs will be fully automated. These estimates are 13 and 48 years respectively ahead of the dates given in last year's survey results.