As the pace of artificial intelligence advancement far exceeds industry expectations, pioneer in AI safety officially announced the establishment of Anthropic Institute, a think tank, on October 11th local time in the United States. The mission of this organization is not to develop more powerful models, but to focus on researching and addressing the deep challenges and serious risks that "powerful AI" may bring to human society.
predicts that artificial intelligence will see breakthroughs in the next two years, possibly reaching the edge of general artificial intelligence (AGI) earlier than most people imagine. To this end, the newly established think tank will focus on the following four key areas:
Employment and Economic Transformation: Studying the reshaping of the global labor market by AI automation and the corresponding response mechanisms.
Threats and Defensive Resilience: Preventing the misuse of AI technology in high-risk fields such as biosecurity and cyberattacks, and enhancing societal defense capabilities.
AI Value Alignment: How to ensure that the decisions of cutting-edge AI systems align with human ethical standards and public interests.
Self-Loop Governance Systems: Exploring transparency and compliance management of AI in self-supervision and self-evolution processes.
The goal of is to break the closed nature within tech giants. It will act as a bridge, sharing with the public and decision-making bodies the real technical challenges encountered during the development of cutting-edge systems, and will deeply collaborate with external research institutions, policymakers, and civil society organizations to jointly develop solutions.
Recently, has performed exceptionally well commercially. There are reports that its , a model under , has added over one million users per day. However, behind the surge in computing power and wealth, has always maintained its "safety-first" corporate identity.
