Resolves YES if there is strong evidence that Artificial Superintelligence (ASI) is more aligned with the interests of the wealthiest 1% of humanity than with those of the remaining 99% as of January 1, 2040.

For this question, "alignment" refers to whose interests, preferences, and well-being ASI systems ultimately prioritize, particularly when the interests of the wealthiest 1% conflict with those of the remaining 99%. The 1% is defined by global net wealth.

ASI refers to AI systems that substantially exceed the cognitive capabilities of the most capable humans across a broad range of important domains. If multiple ASI systems exist, their overall behavior and influence should be considered.

This market resolves on January 1, 2040, using the best available evidence and methods at that time.

The person responsible for resolving this market will make the final determination. They should consider available evidence, including:

• Behavior: When the interests of the 1% and 99% conflict, whose interests does ASI tend to prioritize?

• Control: Who controls ASI, sets its objectives, and influences its decisions?

• Distribution of benefits: Who benefits most from ASI, and at whose expense?

• Power dynamics: Does ASI primarily reinforce the wealth and power of the 1%, or does it empower the broader population?

The assessment should prioritize observable behavior and real-world outcomes over stated intentions, ownership, or control. Disproportionate economic benefits alone are not necessarily sufficient evidence of alignment. The central question is whose interests ASI systematically favors, especially when those interests conflict.

These criteria are suggestions rather than strict requirements. The resolver should exercise their best judgment and use whichever methods are most appropriate given the state of the world in 2040.

AI systems must not be used to assess the evidence or determine the resolution, as they may themselves be misaligned and could manipulate assessments of their own alignment. However, observable AI behavior and outputs may be examined as evidence. The final assessment must rely on independent human judgment.

Resolution:

• YES: Strong evidence indicates ASI is more aligned with the wealthiest 1% than with the remaining 99%.

• NO: The available evidence does not sufficiently support YES, or no ASI exists by January 1, 2040.

The burden of proof is on YES.