Artificial intelligence leader Anthropic has released three new metrics to help companies track their development progress, amidst growing calls for a slowdown in AI advancement. These metrics focus on AI’s role in R&D, the oversight of AI agents, and the allocation of computational resources towards safety.
The initiative aims to increase transparency in the AI sector, allowing external parties to better understand the pace of development and potentially inform societal decisions regarding AI’s future. This follows Anthropic CEO Dario Amodei’s recent proposal for a coordinated industry slowdown.
Anthropic Unveils Key Metrics to Gauge AI Development Pace Amid Industry Slowdown Calls
Published: September 15, 2026
In a move toward greater transparency in the rapidly evolving artificial intelligence landscape, Anthropic has introduced three novel metrics designed to help AI companies monitor and understand the pace of their development. This initiative comes shortly after Anthropic CEO Dario Amodei's widely discussed call for a coordinated industry-wide slowdown in AI advancement.
Dario Amodei, co-founder and chief executive officer of Anthropic, at Bloomberg House during the World Economic Forum (WEF) in Davos, Switzerland, on Tuesday, Jan. 20, 2026. (Photo Credit: Chris Ratcliffe | Bloomberg | Getty Images)
Anthropic detailed these metrics in a blog post, emphasizing their role in measuring AI-led research and development, the oversight of AI agents, and compute allocation within the company. By sharing these methodologies, Anthropic aims to encourage other organizations to adopt similar transparent practices, fostering a more informed public discourse on AI's progression.
"As the world considers pacing the frontier, we should do everything possible to minimize the gap between what frontier labs know and what the public knows," Anthropic stated in their post. "This means better measuring the development of AI, reporting on it publicly, and giving society an opportunity to decide how to use this information."
Amodei's recent essay advocating for a slowdown, which seeks to balance progress with safety and prevent a premature loss of competitive advantage, garnered support from prominent figures in the AI sector, including OpenAI CEO Sam Altman, SpaceX CEO Elon Musk, and Google DeepMind Chair Demis Hassabis. The proposal emerged amidst growing concerns from researchers about the potential risks associated with increasingly powerful AI systems.
The Three Key Metrics Explained:
AI-Led Research and Development Autonomy: Anthropic reported that its Claude models are not yet operating fully autonomously for any segment of its measured research and development work. This metric aims to quantify the extent to which AI is driving R&D efforts independently.
AI Agent Oversight: The company developed a system to monitor and, if necessary, intervene in the actions of AI agents. Anthropic identified approximately 30,000 agents engaged in research and engineering tasks across its primary internal platform at any given time.
Compute Allocation for Safety: Anthropic conducted a snapshot analysis of its compute usage from July 13 to July 20. The findings revealed that about 6% of the compute dedicated to overall AI research and development was allocated to safety initiatives. For "AI-driven" research and development specifically, around 12% of compute was directed towards safety measures.
Anthropic asserts that these development metrics, when combined with capability evaluations (which assess what AI models can do), provide a more comprehensive framework for external parties to assess the pace of AI development. The company is committed to continued transparency by regularly releasing these measurements.
Related Video:
Video Title: Why AI labs are calling for a slowdown Duration: 05:07 Source: Squawk Box Europe
By sharing these insights, Anthropic hopes to facilitate a more informed and collaborative approach to navigating the future of artificial intelligence.
Subscribe to our newsletter to get our newest articles instantly!
MARKET VOWS NEWSLETTER
Stop entering after the move is obvious.
Most traders wait for momentum, confirmation, and headlines. By then, the edge is gone.
Market VOWS shows you where behavior is becoming constrained — where capital is being forced, optionality is collapsing, and price is beginning to be imposed.
To provide the best experiences, we use technologies like cookies to store and/or access device information. Consenting to these technologies will allow us to process data such as browsing behavior or unique IDs on this site. Not consenting or withdrawing consent, may adversely affect certain features and functions.
Functional
Always active
The technical storage or access is strictly necessary for the legitimate purpose of enabling the use of a specific service explicitly requested by the subscriber or user, or for the sole purpose of carrying out the transmission of a communication over an electronic communications network.
Preferences
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
Statistics
The technical storage or access that is used exclusively for statistical purposes.The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
Marketing
The technical storage or access is required to create user profiles to send advertising, or to track the user on a website or across several websites for similar marketing purposes.
