Safety First

The 2023 Expert Survey on Progress in AI, published by researchers from Berkeley, Bonn and Oxford universities found that more than a third of respondents assign at least a 10% chance that advanced Artificial Intelligence could lead to human extinction. Nearer term, assuming humanity survives, 50% of AI experts responding to the survey believe AI will enhance and speed the spread of false information and over 40% are concerned that AI systems could manipulate large-scale public opinion trends. As a result, 34% of AI experts believe safety should be prioritized much more, and 36% believe it should be prioritized more.

Source: AI Impacts, January 2024