Smart contracts and kill switches in the EU Data Act.
On 27 November the European Council approved the controversial European Data Act, aimed at “establishing harmonized rules on making available to the user of a product or service the data generated by the use of the product itself or a related service (omissis )“.
The new regulation, which embraces a very broad scope of technical regulatory aspects, dedicates ample space to the topic of smart contracts.
In the blockchain context, smart contracts are conditional transactions with respect to the occurrence of some event, whether predefined (e.g., a certain date), or deriving from external input (e.g., reaching price thresholds of a certain good) according to what is rendered available to the blockchain from a specific category of agents, the Oracles.
The central point, which slows down the diffusion of smart contracts, is that a smart contract cannot be interrupted in its execution by the blockchain. A smart contract is, in fact, “unstoppable”, whatever it is programmed to do.
Smart contracts are implemented on some advanced second generation blockchains, including in particular Ethereum, Cardano and Solana, but not on the first and best-known one, the Bitcoin blockchain.
In preamble no. 104 of the Data Act, the concept of a smart contract termination is introduced by mutual agreement of the contracting parties. “The notion of ‘smart contract’ in this Regulation is technologically neutral. Smart contracts can, for example, be connected to an electronic ledger. The essential requirements should apply only to the vendors of smart contracts, although not where they develop smart contracts in-house exclusively for internal use. The essential requirement to ensure that smart contracts can be interrupted and terminated implies mutual consent by the parties to the data sharing greement. The applicability of the relevant rules of civil, contractual and consumer protection law to data sharing agreements remains or should remain unaffected by the use of smart contracts for the automated execution of such agreements.”
The concept of an agreed upon smart contracts termination is referred to in technical jargon as a kill switch. The logic is to provide the interrupt functionality of a smart contract, which is currently not guaranteed, and which risks being potentially devastating in the event of coding errors or cyber attacks. An incorrect program that cannot be blocked because it runs on a distributed blockchain risks causing serious damage to the parties involved. Maybe due to a trivial error, a zero more or less in an amount that no one noticed before the launch on the BC.
Article 36, first 5 paragraphs, goes into detail about the regulatory provisions of the Data Act regarding kill switches: “the seller of applications that use smart contracts or, in his absence, the person whose commercial, entrepreneurial or professional activity involves the implementation of smart contracts for others in the context of the execution of a data provision agreement, or part thereof, ensures that such smart contracts comply with the following essential requirements:
a) (omitted..)
b) safe termination and interruption, to ensure that a mechanism exists to terminate the continued execution of transactions and that the smart contract includes internal functions which can reset or instruct the contract to stop or interrupt the operation, in particular to avoid future accidental executions.“
The related Generative Artificial Intelligence question is posted in the dedicated gallery: What are the mechanisms for implementing kill switches?