In the past six months, it seems the concepts of "AI" and "data-driven digital transformation" have become so blurred as to be indistinguishable. The gist of my argument is that any company or organisation that wants to sound like it's "on point" technically is using the term AI when what it really means is "using digital".

The latest example is The Olympics, which seems to be hailing Paris 2024 as the first AI Olympics. The trouble is, much of what they are calling out as AI is really not.

Among the initiatives labelled AI are the following:

​

• A chat service for athletes to get access to important information to "provide easy and quick answers to Frequently Asked Questions on topics such as social media guidelines, anti-doping rules, and Rule 50 regulations." I struggle to see how AI is critical to a searchable archive of FAQs

• Efficient planning through Digital Twinning - this enables the Olympics to create digital copies of buildings to know where to fix TV cameras and the like. Once again - a smart use of technology but hardly new and it's difficult to understand the role AI is playing here. Digital Twinning has been around for more than a decade and deployed in thousands of scenarios without the need for AI

• "The energy consumption at Paris 2024 will be monitored in real time, and the captured data will be used to inform future planning". Smart application of AI should enable power to optimise performance in real time against ever changing criteria to deliver a "most optimal" solution - rather than gathering data for future planning?

​​

To be completely fair to the IOC, the post does openly admit that "We are taking a measured approach for now, to test and evaluate how AI can be used." I also appreciate that the Olympics has many "technology partners" (read "sponsors") that will be salivating to get their dollar's worth of kudos from the Olympics, associating them with the "next big thing". Despite all this, it still feels like this is an attempt to conjure AI into the conversation rather than apply it effectively. We'll see more of this I'm sure in the coming months as the galloping herd of technology brands slavishly follows the zeitgeist.

​

And to counter the accusation "Well what would you do better," for my money, AI could be valuably deployed in the Olympics to do the following:

​

• Optimise the mobile and Wi-Fi networks to deliver the most optimal performance based on visitor numbers, location of fans and prioritisation of officials

• Monitor and optimise traffic around a notoriously difficult city, delivering smoother visitor travel and real time prioritisation for emergency services

• Analyse athlete performance against historical data to seek anomalies in real time and use this to prioritise drug testing

Another AI story that caught my attention this week related to OpenAI and its ability to burn through cash at an astonishing rate (along the lines of - wait for it - $8.5 billion a year). It prompted me to write a blog speculating whether investors will ever see a return on investment or if other companies will benefit instead. You can read it here.