Smart contracts, which are self-executing and self-enforcing, could bring greater reliability and efficiency to business processes (in the realm of payments and cash management, the most probable use is in trade finance). They raise the possibility that companies would no longer need to revert to courts of law to have contracts enforced, because the smart contract would automatically ensure that A pays B, on X date provided conditions Y and Z are met. This makes it sound simple and easy, but there are downsides to relying on computer coding to decide the outcome of legal agreements.

A paper (Contracts Ex Machina) by two Wharton professors of legal studies and business ethics, Kevin Werbach and Nico Cornell, looks at the potential and the drawbacks of smart contracts. They say: “The reality is, even though we think machines can render contracts effectively, there are lots of situations where they cannot.”

So what are some of the issues that businesses should be aware of when considering the use of smart contracts?

1. Some contracts contain terms that machines might not recognise. Werbach and Cornell use the example of contracts with clauses referring to the “use of best efforts”, which is difficult to translate into computer code, since there is an element of human judgement in considering whether one party in the contract has used its 'best efforts'.
2. Smart contracts don't provide the possibility of re-negotiation. Traditional contracts allow the parties to reconsider if the outcome of the contract is no longer considered to be of mutual benefit. Werbach says: “The smart contract is not smart enough to do that, unless the parties build into the smart contract code at the beginning the possibility for that modification.”
3. Smart contracts are based on computer coding, so are subject to bugs. If the computer coding leads to an outcome that wasn't intended by the humans who entered into the contract or there is a misinterpretation or misunderstanding of the contract's terms, there is no human arbiter or judge to interpret and enforce or annul the contract.
4. Anything code-based could be subject to an attack by hackers, with the possibility of the smart contract's code being compromised and changed to make a fraudulent payment to the wrong account. In such cases, the computer has not way of knowing that the fraud is occurring because it has been coded to accept the payment. The fraud can only be detected if a human goes and checks. And of course the whole point of smart contracts is that they don't need to be monitored by humans.

Despite the concerns outlined in their paper, Werbach is clear that we'll be seeing and using smart contracts a lot more in future. He says: “The opportunity here is just immense... We clearly are going to see smart contracts get used more and more widely.”