Market desk Bitcoin Ethereum Altcoins DeFi Stablecoins Markets & Trading

Vitalik Buterin Endorses AI Model's Decision to Limit Hacking Capabilities

Ethereum co-founder Vitalik Buterin has backed Moonshot AI's decision to remove top-end hacking skills from its Kimi K3.1 Pro model, arguing that limiting AI capabilities is preferable to restricting user access.
1 day ago 2 views
Vitalik Buterin Endorses AI Model's Decision to Limit Hacking Capabilities

Ethereum co-founder Vitalik Buterin has endorsed Moonshot AI's decision to deliberately remove advanced hacking capabilities from its new Kimi K3.1 Pro model. According to Buterin, restricting what AI can do is less harmful than restricting who can use it.

Moonshot, a Beijing-based AI startup, removed cyber-related data from its training corpus and reinforcement learning environments while preserving general coding tasks. The company reported that the filtered model solved 81% of tasks on its internal exploit benchmark, compared to 90.5% for an unfiltered version.

Why Capability Limits Matter More Than User Restrictions

Buterin argued that discriminating against use cases is significantly less damaging than discriminating against users. He emphasized that developers need time to harden systems, check legacy code, and verify new software before AI-powered attacks reach full capability.

The concern reflects existing vulnerabilities. Earlier this year, Binance Research found that AI exploits smart contracts roughly twice as well as it detects flaws.

Open-Source Release and Competitive Pressures

Moonshot plans to release the model's open weights by November 4, allowing anyone to download and run it on their own hardware. However, other open-source models are approaching similar performance levels. DeepSeek's rival open model V4.1-Flash already scored 85.7% on the same benchmark.

Buterin acknowledged that filtering will not permanently prevent the development of such capabilities. He framed AI safety as a race for time rather than a permanent block, favoring defenders who adopt redundant crypto security safeguards before attackers advance their techniques.

Alternative Scenarios and Surveillance Risks

Buterin warned that if this initial period of AI development proceeds poorly, it could result in locked-down models, Know Your Customer checks on all systems, and intrusive surveillance. In such a scenario, AI access could become dependent on credentials and nationality, creating different classes of users with varying levels of access.

Market snapshot

Top cryptocurrency prices

Explore all prices
BitcoinBTC $79,987.88-0.47% EthereumETH $2,317.58-0.46% Tether USDUSDT $0.9996-0.04% BNBBNB $740.20+0.05% XRPXRP $1.30+1.22% USDCUSDC $1.000.00% SolanaSOL $100.18-1.11% TRONTRX $0.3281-0.91% HyperliquidHYPE $77.88+1.19% ZcashZEC $1,253.33+2.16%
Prices by Coinranking. Informational only.