Anthropic announced that the Claude AI model significantly aids in the development of its successor. This news was released on Thursday and highlights Claude's role in the company's research and development.
Claude's Role in Research and Development
Claude leads 26% of Anthropic's research and development, which the company states means that this model can perform more of a task "from start to finish based on a high-level instruction," while still being under human supervision. This model does not yet operate fully autonomously.
Read more: Joe Rogan spoke about AI governance as a way to end wars
However, about 90% of the company's research and development is done in "collaboration" with Claude. Anthropic has stated that this model can "perform large parts of the work under close human guidance."
Concerns About AI Safety and Development
This announcement comes as some prominent figures in the AI field, led by Dario Amodei, CEO of Anthropic, are calling for a slowdown in technology development due to safety concerns. Anthropic emphasized that "we must do everything we can to minimize the gap between what advanced labs know and what the general public knows."
The company has also stressed the need for other AI developers to share similar metrics and encouraged the use of a public method for comparing these numbers.
Anthropic revealed details of the measures it is taking to monitor its models. In August, about 30,000 agents were involved in research and engineering at the company. These monitoring measures are essential for detecting agents' misconduct through oversight systems. Anthropic has recently committed to placing independent evaluators to oversee its safety efforts within the company.
Last week, an Anthropic researcher resigned, seriously warning about the threats that AI technology poses to humanity. Amodei, Sam Altman, CEO of OpenAI, Elon Musk, and other tech leaders have supported the idea of slowing down development, while some tech leaders and Donald Trump have opposed this notion.
Read more: Two-thirds of Americans take AI risks seriously · Mark Zuckerberg distanced himself from the call for a coordinated slowdown in AI development
Source: abcnews.com
