Analysing the Impact of AI on Business with David Skerrett & James Lawson

In November, our ongoing AI webinar series returned to assess the impact AI is having and will continue to have on business. Webinar chair David Skerrett and guest speaker James Lawson, an AI Evangelist for DataRobot, discussed how businesses are using AI to accelerate their processes and unlock greater revenue potential.

Our returning and popular segment The Good, the Bad and the Ugly also saw David and James open the floor to our audience. They introduced a variety of AI use cases and briefly discussed their impact before asking the audience to cast their vote on whether the use cases were good, bad or downright ugly.

Here’s a quick round up of the third webinar in our going All Things AI series, brought to you by sponsors DataRobot. (More on who DataRobot are later!)


Introducing AI in Business

To kick the webinar off,’s Craig McCartney introduced David Skerrett, our webinar series chair. 

David has had his finger on the pulse of AI in business for several years now, and he believes that we’re living in a time of great innovation. In fact, AI is significantly impacting start-up and enterprise businesses globally. These companies are now able to make smarter decisions, enhance processes and make more accurate predictions, streamline manual tasks, tackle fraud and create a more compelling and personal customer experience. 

David noted that AI is driving global business evolution. That’s because 84% of businesses say AI enables or sustains competitive advantage, and 83% say AI is a strategic priority for their business today. For those enterprises that do incorporate AI tools, a 120% increase in economic value by 2030 is the prize. Other estimates suggest that AI will contribute $15.7 trillion to the global economy by 2030.

If these predictions weren’t enough to convince businesses of the imperative to adopt AI, David had news… AI is driving expectation inflation. Now, customers and businesses alike expect better and faster services. Think about how delivery times have changed since a decade ago, when a three-day delivery was considered good: today, Amazon offers same-day delivery, and customers expect it. In short, AI isn’t a nice-to-have, it’s a business necessity. 


Extracting Value from AI

The need to adopt AI is now a reality that businesses can’t ignore, but it isn’t just because customers and businesses expect AI-driven solutions now.

It’s because of the extreme revenue opportunities AI brings with it.

Business leaders are monetising AI at an unprecedented rate. The value extraction is about to skyrocket across business functions, as David demonstrated with the following statistics: 

David wrapped up his part of the webinar by saying that although businesses may want to adopt AI, the only way to do so successfully is with this handy equation: 

People + Technology + Trust = Transformation Success


AI isn’t here to replace people, it’s here to augment the processes we already do by automating the tedious tasks so people can do more of the complicated tasks that unlock new business ventures.

The only way to get people on board is by gaining their trust through use cases that show their jobs aren’t at risk.

This is the path to transformation success, and here was a great time to hand over to our guest speaker, James Lawson.


Data Is the New Oil

big tech companies on oil rigs david parkins economist
Data is the new oil. Photo credited to The Economist.


James Lawson took over the discussion with the well-known Economist front cover: The World’s Most Valuable Resource, Data. Data is the new oil, and that’s primarily because it is the fuel that powers AI, which is what’s creating new opportunities for businesses.

James is an AI evangelist, and it’s a great time to be one. As James pointed out, companies are mentioning AI in earnings calls more and more, with massive rises since 2016. 

It’s no wonder that more companies are mentioning AI, given that first-mover advantages are huge. Those companies that adopt AI first are already reaping massive value from AI because, as Kai-Fu Lee highlighted, “The positive-feedback loop generated by increasing amounts of data means that AI-driven industries naturally tend toward monopoly.”  

As companies accrue data, they tend to accrue more, and this data powers greater AI processes that in turn power better business practices and generate more revenue… and data. It means that early adopters are able to better serve the market and that laggards simply can’t keep up because they don’t have the time to accrue as much data.  

As the Harvard Business Review suggests, “The [AI] winners may take all and late adopters may never catch up.”


AI Is Difficult to Adopt

Although there’s a clear need for businesses to adopt AI to remain competitive, AI isn’t a walk in the park. It is difficult to adopt and many organisations encounter issues when they underestimate AI.

  • 96% of organisations experience problems with model development
  • 90% of organisations cite obstacles to moving AI models into production
  • Just 1% of organisations monitor their AI and ML assets once they’re liveFrom build to deployment and beyond, organisations struggle with AI, stumbling at obstacles every step of the way.

Whether it’s that organisations fail to deal with hundreds of potential use cases for AI or resource projects effectively, AI adoption proves an issue.

That’s why James was keen to outline a maturity index for organisations. From Sceptic to AI-Driven and everything in between, James charts the typical route to AI maturity he has seen in organisations while working at DataRobot. He talked about how organisations first adopt AI because they have to, but as they learn more about the value of AI and the need to accrue more data, their attitudes change.


When AI Is Done Right, It Has a Huge Impact

For those who can adopt AI right, the gains can be mind-blowing. DataRobot has helped clients to adopt AI successfully and James talked us through some of the greatest successes.

  • A healthcare provider saved $18m by improving nurse staffing and reducing patient stays in a safe, reliable way
  • A retailer made $200m by forecasting use cases more accurately
  • DataRobot helped generate $400m worth of value for the top 5 global banks across multiple use cases

James’ lesson is clear:

Anybody not adopting AI now will fall behind.

James has seen this firsthand. Years ago, organisations wanted to review multiple proof of concepts before they invested in AI. Nowadays, businesses are scared they’ll be left behind and simply want AI implementation done, so settle on the first concept that looks good!

Among the fastest moving organisations are those in the insurance, banking, telcos and retail sectors, but these industries are quickly being followed by pharmaceuticals and CPG manufacturers.

Sadly, James has seen that the public sector is lagging behind, though the military is investing in some technologies.


The Good, the Bad and the Ugly

James concluded his section of the webinar by saying that anybody who doesn’t adopt AI now won’t remain competitive and will be left behind. The onus is on organisations like yours to think creatively about how you might adopt and deploy AI to extract the most value. 

With that said, our speakers had collected six interesting use cases from around the world. We opened the floor to our audience to see whether these were Good, Bad or Ugly. 

Met Police & Facebook Collaboration

From October 2019, London Metropolitan Police began providing Facebook with video footage of its firearms training from the perspective of officers. The hope is that this will help Facebook to develop an AI solution that will detect when someone is livestreaming footage of a firearms attack. This could mean that livestreams of shootings are taken down in real-time and Facebook could even notify police of an attack early on. 

Good 6%
Bad 44%
Ugly 50%

Tackling the Water Challenge in Africa

Access to water is a human right, but for many people in the world it’s simply inaccessible. Nearly one billion people rely on rural water points, but these typically break every three years – 25% don’t work today.  

Global Water Challenge decided to develop an AI solution that would answer questions like which water point would break next, so developed the first harmonised database of water points from around the world. Following a successful pilot in Sierra Leone, the company rolled the solution out to 13 countries to identify water point breaks quickly and enhance the water system generally. 

Good 100%
Bad 0%
Ugly 0%

Microsoft AI Takes on Beach Pollution

An unimaginable 5 trillion pieces of plastic weighing more than 250,000 tons are floating in the world’s oceans. Unsurprisingly, beach litter is therefore fast becoming an increasing health and environmental hazard. But in 2017, the UN said that it can’t improve what it doesn’t measure.  

An initiative in New Zealand took this as a call to arms. It began using AI to categorise, gather, visualise and understand plastics in the ocean. The solution from Sustainable Coastlines enables citizens to get involved in clearing beaches of plastic. 

Good 100%
Bad 0%
Ugly 0%

Kroger Predicts Apples & Oranges

Kroger is the second largest retailer in the US. As such, Kroeger knows a thing or two about how demand can vary significantly based on date, time, location, events and more. But predicting what products every supermarket will need and when is still extremely difficult.  

To make it easier to get the right products to the right stores at the right time, Kroger is developing an entire demand forecasting system to improve operations. It looks at sourcing/manufacturing for each item/size/colour/location to help customers get the goods they want while helping the environment by reducing waste. 

Good 74%
Bad 16%
Ugly 10%

Jaguar Gives Eyes to Its Self-Driving Cars

63% of pedestrians and cyclists feeling less safe at the prospect of autonomous vehicles on the road. To help, Jaguar Land Rover has tested a novel way to make humans feel safer around self-driving cars: giving cars virtual eyes. 

In trial studies, JLR put the virtual eyes on self-driving pods to measure how much people trusted autonomous vehicles. The eyes made direct eye contact with pedestrians to signal intent. The aim was to see if pedestrians trusted the vehicles more if they had an intent signal from the car to tell them it was safe to cross. If ongoing studies turn out positive, we could see self-driving cars with eyes before long… 

Good 29%
Bad 7%
Ugly 64%

Detecting Fraud at American Fidelity

Fraud is a pervasive problem in the insurance industry. From claims for properties that don’t exist or were intentionally damaged, or fraudulently claiming on theft, medical bills and more, insurance companies lose millions every year.  

American Fidelity tried to curb this issue with AI. It developed a model that was able to classify cases as fraud based on past data and identify fraudulent behaviour earlier. They enhanced the AI with anomaly detection processes to spot outliers. The positive aspect of this is that it should cut fraud. The downside is that not all anomalies are fraudulent, and cases that match historical fraud cases aren’t always fake.

Good 77%
Bad 8%
Ugly 15%

Want to Learn More?

Ready to learn more about what David and James talked about in this webinar? You can view the full recording of the webinar to glean even more insights from our expert speakers. 

We’d like to give another shout out to our sponsors, DataRobot, for making this webinar possible. We hope you learned a lot about the innovative ways that AI is affecting business operations around the world. 

Read More News Articles

Introducing Our Sponsors, DataRobot

This webinar was brought to you by DataRobot. Boasting the world’s only trusted enterprise AI platform, DataRobot enables organisations to leverage the transformational power of AI and turn data into value. Head to DataRobot today for a free demo of their innovative platform.