Key facts
- An Anthropic employee expressed confusion and uncertainty about their job due to AI advancements.
- The employee stated that AI automation makes them question the relevance of their work.
- Over 80% of code merged into Anthropic's codebase was authored by its AI model, Claude.
- Anthropic has proposed a coordinated slowdown or pause in AI development among frontier labs.
- The Anthropic Institute plans to study systems needed to support a coordinated slowdown.
An employee at AI startup Anthropic has expressed significant confusion and existential turmoil regarding their job due to rapid AI advancements. In a company blog post, the employee stated, "On days where everything works well, I can't help but think nothing I do matters, everything is automated and better and faster than I ever will be." They added, "But then there are days where everything breaks and I don't understand why and I realize I have no idea what I've been up to anymore." This sentiment reflects broader concerns about AI's impact on the workforce, with AI models like Anthropic's Claude capable of generating code and performing complex tasks. One employee noted they hadn't written code themselves in about five months, while another predicted AI-generated code would surpass human-written code within a year. Despite Claude's capabilities in engineering problems and research, Anthropic acknowledges performance gaps in its ability to exercise judgment in choosing goals. As of May, more than 80% of the code merged into Anthropic's codebase was written by Claude. Anthropic has proposed a coordinated slowdown or pause in AI development among frontier labs, warning that AI systems may soon be capable of designing and training their own successors without human involvement. This proposal, detailed by the Anthropic Institute, aims to allow societal structures and alignment research to keep pace with AI's recursive self-improvement capabilities. The company has confidentially filed for a U.S. IPO and was last valued at $61 billion.