In Anthropic’s new documents, it reiterated promises it has made in the past to develop AI in a way that is beneficial to people, while at the same time warning investors of its dangers.
Anthropic detailed losses in its financial filings that exceeded its revenue for the first half of the year, but said it would often make forward investments “that may seem irresponsibly large at first glance”.
“In general, we will aim to be bold but sane, and to win trust with a track record of successful bets,” the tech giant said.
Its success depends largely on its ability to develop the most advanced AI models, Anthropic said, adding, “any failure to do so could adversely affect our business”.
Across about 90 pages detailing possible risk factors, Anthropic set out the potential threats to its ability to safely and competitively develop AI.
Anthropic warned it relies on a cluster of cloud computing providers, Taiwanese chipmakers already facing huge demand and data centres concentrated in the US.
It also warned of “regulatory action arising from harms associated with our AI models”
However, Anthropic also used the filing to laud the potential benefits of the tech itself, envisioning it unlocking “millions of virtual agents with the ability to revolutionise every area of human endeavour” across a range of fields.
“We believe that AI is one of the most consequential technologies in human history,” the firm said.
Anthropic said AI would also “profoundly accelerate the rate of progress in biology and medicine”, potentially finding cures to diseases including cancer and Alzheimer’s.
The firm added it could also establish “sustained acceleration in economic growth” by boosting productivity.