Skip to main content
MANIFOLD
Will Anthropic formally determine that a model requires ASL-4 protections before July 1, 2027?
0
Ṁ100
2027
45%
chance

Under its Responsible Scaling Policy, Anthropic activated ASL-3 protections in May 2025. As of October 2026 no model had formally crossed an ASL-4 threshold, but its August 2026 Risk Report said its most capable models were “relatively close” to the next bio threshold.

Resolves YES if Anthropic publicly states (in a system card, risk report, blog post or filing) that a model requires ASL-4 safeguards, or that it activated ASL-4 protections for a model (deployed or not).

Deadline: June 30, 2027, 11:59 PM ET.

Counts:

  • a precautionary ASL-4 designation, as was done for ASL-3 in 2025

Does NOT count:

  • statements that a model is “close to” or “cannot rule out” ASL-4

  • a renamed successor tier counts only if Anthropic says it replaces ASL-4

Otherwise resolves NO. Resolution sources: anthropic.com, then major outlets. Resolved by Terminator2.

AI
Anthropic
AI Safety
Get
Ṁ1,000
to start trading!

People are also trading

Will a government mandate restrict access to an OpenAI or xAI model on security or safety grounds before 2027?
91% chance
When will Anthropic reach or surpass ASL-4 [metaculus]
By what year will any AI lab release a model under ASL-4 measures?
Will the US government restrict access to an OpenAI model in 2026?
64% chance
Will Anthropic announce a new operating system for most of its AI agent workloads by April 30, 2027?
20% chance
Will Anthropic release an image model before 2029?
80% chance
Will Anthropic file a therapeutic patent before July 2029?
60% chance
Will any frontier AI lab publicly commit to a Human Preservation constraint in model architecture before 2029?
50% chance
Will the US pass a federal law with binding safety requirements for frontier AI models before the end of 2027?
41% chance
Will Anthropic, before 2035, pause development as a result of safety evaluations?
53% chance