AI gone wrong…

As AI’s presence in the commercial world increases, a recent Canadian tribunal has many questioning how reliable AI is when it comes to sales, negotiation and customer interaction.

AI is being utilised across industries to assist businesses in automating tasks, enhancing decision-making with data analytics, personalising marketing campaigns, optimising supply chain logistics, predicting trends for strategic planning and improving customer service through chatbots. The latter is in great debate at present.

Air Canada has been ordered to pay compensation to a customer who was informed by a chatbot that they could purchase a ticket and later claim a partial refund due to bereavement. There is much debate about who is responsible when chatbots get it wrong. Air Canada argued that the chatbot is a “separate legal entity”. The court ruled that Air Canada must take responsibility for any refunds promised by the AI chatbots on their website.

Although the terms, timing and process in the refund claim are where the issues in this particular situation arose, it does raise the question of how reliable AI is when negotiating the finer details of a sale/contract. Who should be held responsible when AI gets it wrong and what measures are in place to ensure AI isn’t misinforming customers or businesses?

As more and more companies begin to integrate AI technology into their front and back-end processes there must not only be an awareness of the value AI brings but also the rate at which it is changing. Companies will need to ensure their positioning and safeguarding against potential issues evolve at the speed of AI’s development.

This brings us to more rampant debates at present – should AI be classed as a separate legal entity? Will we humans be able to keep track/in control of AI? Will there come a time when we have to consider the sentience of AI? Will AI be held responsible for misinformation… and how will that put businesses in a position of strength or weakness over the coming years...

When it comes to negotiating, AI now has decision support systems and data analysis abilities that can certainly help with preparation. Decision support systems can help you to analyse potential negotiation scenarios, create wish and concession lists or consider constraints that may affect the negotiation. All of these can be helpful in preparation for a negotiation when time is of the essence. AI analysing large volumes of data, market trends, patterns and similar deals can also help with strategic preparation. So, we certainly aren’t suggesting that AI hasn’t got a place in negotiation, we merely need to understand how to best utilise it.