Tesco could use AI to tell you off if your shopping basket is too unhealthy - how it could work
and on Freeview 262 or Freely 565
- Tesco chief executive floated the idea of using AI to ‘nudge’ Clubcard members.
- It could encourage shoppers to replace items in their basket with healthier alternatives.
- He also claims it could bring “your shopping bill down” and “reduce waste”.
If you go down to Tesco for your weekly shop, you may be in for a big surprise.
The supermarket could - in one possible future - use artificial intelligence (AI) to analyse Clubcard members' latest shop and give them a telling off if they have too many unhealthy items in their basket. So perhaps in preparation for that day it may be prudent to think twice before reaching for that packet of fancy biscuits that have been seductively calling your name - or, in the words of internet favourite Limmy, ‘don’t back down, double down’.
Advertisement
Hide AdAdvertisement
Hide AdIt would certainly be among the most novel uses of AI - but while it is a grand claim it does leave plenty of questions. Not least of all when it comes to whether this is an ethical use of customer’s data.
But what exactly has been said? Let’s find out:
What has the Tesco CEO said about AI?
Speaking at the FT Future of Retail Conference recently, the CEO of Tesco, Ken Murphy, made his claims about how artificial intelligence could be used to “nudge” customers out of unhealthy eating habits. The Mirror reports that he said: “I can see it nudging you, saying: ‘look, I’ve noticed over time that in your shopping basket your sodium salt content is 250% of your daily recommended allowance. I would recommend you substitute this, this and this for lower sodium products to improve your heart health’.”
Tesco currently has over 20 million Clubcard members, so that would certainly be a lot of nudges to send out. The CEO was not firm on details, meaning it probably isn’t something to expect when you shop at the supermarket in the near future.
It would certainly require a lot of work, not least of all how it would send the “nudges”. And it could make self-checkout tills a lot more chaotic if customers are rushing around attempting to replace items they’ve already scanned for the AI-suggested substitutes - perhaps it would only work with ‘scan as you shop’ shoppers who could more easily adjust items in a basket.
Advertisement
Hide AdAdvertisement
Hide AdBut what about customer data?
During the conference, Mr Murphy added: “It will completely revolutionise how customers interact with retailers. It can help to bring your shopping bill down, reduce waste and improve the outcome and power of that Clubcard.”
Addressing concerns that customers may have about their data being used in such a way, he said: “For me as a consumer, I don’t mind giving up that data if I get the right kind of experience back.”
The Mirror reports that Tesco and Sainsbury’s are said to earn around £300 million a year from selling customer data acquired via their own loyalty schemes (Clubcard and Nectar card).
What does ‘nudging’ mean?
You might have heard the term ‘nudge’ during the pandemic in particular. The chances are you are far more likely to be familiar with the word in its everyday use: gently pushing someone to get past or to get their attention.
Advertisement
Hide AdAdvertisement
Hide AdHowever it does have a public policy meaning - and this is the one being referred to when it comes to AI ‘nudging’ customers to eat healthier. A nudge in this sense means: interventions that preserve freedom of choice, but that also steer people in certain directions.
The UK government set up the Behavioural Insights team, popularly known as the Nudge Unit, in 2010. It played an important role in formulating the response to the coronavirus pandemic.
Would you be comfortable with Tesco or other supermarkets using AI to ‘nudge’ you into making healthier food choices while in a store. Share your thought by emailing our tech writer: [email protected].
Comment Guidelines
National World encourages reader discussion on our stories. User feedback, insights and back-and-forth exchanges add a rich layer of context to reporting. Please review our Community Guidelines before commenting.