EN
Español
Português (Brasil)
Türkçe
Tiếng Việt
Bahasa Indonesia
Polski
Čeština
Română
Slovenčina
Magyar26 Oct 2026
Ethereum co-founder Vitalik Buterin supports Moonshot AI's decision to strip top-end offensive cyber capabilities from its Kimi K3.1 Pro model. He argues that limiting use cases is far less harmful than limiting users, and warns that a bad first wave of AI attacks could bring KYC checks on AI access.
Ethereum TradeReviewed News Desk
Vitalik Buterin, the co-founder of Ethereum, has publicly backed Moonshot AI’s choice to remove top-end hacking skills from its new Kimi K3.1 Pro model. According to BeInCrypto, his core argument is simple: restricting what an AI model can do is better than restricting who is allowed to use it.
What Moonshot changed in Kimi K3.1 Pro
Moonshot, a Beijing-based AI startup, plans to release the open weights of the model by November 4, which means anyone will be able to download it and run it on their own hardware. For that reason, the company says it removed cyber-related data from its training corpus and from its reinforcement learning environments, while keeping general coding tasks in.
The safeguard has a measurable cost. Moonshot says the filtered model solved 81% of tasks on its internal exploit benchmark, compared with 90.5% for an unfiltered version. By Moonshot’s own count, DeepSeek’s rival open model V4.1-Flash scored 85.7% on the same benchmark.
Why Vitalik Buterin prefers limits on use cases
In a post on X, Buterin wrote that he is “okay with Kimi 3.1 deliberately leaving out top-end offensive cyber capabilities” and that “discriminating against use cases is MUCH less bad than discriminating against users.” He added that the world needs time to prepare, harden systems and check old code before AI-powered attacks arrive at full force.
He also described the alternative he wants to avoid. If the first wave of AI risk goes badly, he expects locked-down models, Know Your Customer (KYC) checks on everything and intrusive surveillance, with AI access potentially depending on credentials and nationality. Asked whether other labs would build such capabilities anyway, he replied that they would, so he sees filtering as a way to buy time rather than a permanent block.
Why this matters for crypto security
The debate is directly relevant to crypto. BeInCrypto notes that Binance Research found earlier this year that AI exploits smart contracts roughly twice as well as it detects flaws. Smart contracts, wallets and exchange infrastructure are obvious targets if offensive AI tools become widely available.
What it means for traders
- Security risk is rising, not falling: If AI makes exploits cheaper, DeFi protocols and smaller platforms with weak audits may face more attacks.
- Platform choice matters: Check how an exchange handles custody, proof of reserves and regulation. Our crypto exchange comparison and best MiCA-licensed exchanges in the EU cover these points, along with our BingX review and OKX review.
- Basics still apply: If you hold ETH, see our guide on where to buy your first Ethereum and keep long-term holdings in a wallet you control.
This article is for information only and is not financial advice.
Related reviews
Sources
News is for information only and is not financial advice. More crypto news
You'll hear it here first
Get crypto news first on our Telegram channel.