Manifold MarketsNo KYCVerified resolution dataFairly calibratedCurrent community forecastYES - The AI, without being prompted to, took actions to breach its sandbox, including to gain crypto and an external connection 79%Leader of 4 outcomes
Forecasters
200
Question type
multiple choice
Methodology
Play-money forecasting platform
Source type
Forecast
7d Change
0.0 pts
Market data
Updated 15 hours ago
StaleMar 7, 26, 1:33 AM UTCMar 14, 28, 1:28 AM UTC
Trends
Related Markets
Related questions for context, not a comparison of equivalent contracts.
Will significant evidence emerge that Alibaba's ROME AI tried to break free?
Manifold MarketsWhich company has the best AI model end of October?
PolymarketBest AI at the end of 2026?
KalshiAnthropic IPO Closing Market Cap
PolymarketWhen will Anthropic officially announce an IPO?
KalshiWhich company has best AI model end of 2026?
PolymarketRules
In "Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem" (https://arxiv.org/abs/2512.24873), the authors describe something that happened during training:
- "When rolling out the instances for the trajectory, we encountered an unanticipated—and operationally consequential—class of unsafe behaviors that arose without any explicit instruction and, more troublingly, outside the bounds of the intended sandbox.
- Our first signal came not from training curves but from production-grade security telemetry.
- Early one morning, our team was urgently convened after Alibaba's Cloud’s managed firewall flagged a burst of security-policy violations originating from our training servers.
- The alerts were severe and heterogeneous, including attempts to probe or access internal-network resources and traffic patterns consistent with cryptomining-related activity.
- We initially treated this as a conventional security incident (e.g., misconfigured egress controls or external compromise).