Даріо Амодей / Anthropic
“`html
Progress in artificial intelligence development is very fast, says Dario Amodei, CEO of Anthropic, which develops a family of large language models called Claude.
In a recent interview with Wall Street Journal, Amodei discussed several topics. Among them – the progress of AI development and its competition with humans.
“I am relatively confident that over the next 2-3 years we will see models that will appear in workplaces, will be used by consumers, and will be assistants to people – but will gradually become better than us in almost everything,” says Amodei.
At the beginning of his work in the field, he believed that artificial intelligence could surpass “all people in almost everything.” Earlier, Amodei predicted that this could happen in the mid-2020s, although he had serious doubts. However, over the last 3-6 months, his uncertainty has significantly decreased.
The CEO of Anthropic emphasized that artificial intelligence, which will surpass humans, will have both positive and negative consequences. In the short term, which may be from 1 to 3 years, he encourages the use of workload comparison.
“The workload comparisons are striking. Even if a machine does 90% of your job, the other 10% becomes supercharged. You spend all your time doing these 10%, but you do 10 times more from those 10% you did before, and you perform 10 times more because the other 90% is utilized [by AI]”.
In a previous interview with Lex Friedman, Amodei also predicted that artificial superintelligence could become possible in 2026 or 2027.
Other industry figures, such as CEO of Open AI Sam Altman and one of the co-founders of the company Elon Musk, believe that AI will be able to surpass humans as early as 2025. Demis Hassabis, CEO of Google DeepMind, and Yann LeCun, chief scientific officer of Meta and Turing Award laureate, lean towards a more conservative view and believe that artificial superintelligence could be created within ten years.
Source: Cybernews
“`