Grid

Chapter 2

AI & Automation

AI & Automation

AI & Automation

Tuesday 6 February 2024

Tuesday 6 February 2024

Jas Bassi

Gateley

Reuben Boughton

GW

James Hobbs

HSBC

Mike Lewis

WMGC

Mike Morgan

Hays

Vinay Parmar

Curo

Dr Sudheer Parwana

PwC

Prof. Abdul Hamid Sadka

Aston Uni

Mark Simpson

GW

Peter Wainwright

ERIKS

Birmingham Tech Leaders: Conversation

Birmingham Tech Leaders: Conversation

Our tech leaders spent 20 minutes discussing the role that AI & Automation holds in their enterprises. And the best and worst use cases that can help us all drive business efficiencies and IT innovation.


We’ve distilled some of the conversation and pointed you in the direction of some fascinating use cases and studies.

Sectors most affected by AI / Deep Fakes & Plausible Deniability

Sectors most affected by AI / Deep Fakes & Plausible Deniability

Sectors most affected by AI / Deep Fakes & Plausible Deniability

Greg Telford

Dr Sudheer Parwana

PwC

“A few years ago, do you remember those Google deep minds? They had this Chinese game called “GO”. It's a very strategic game and very difficult to codify. And it actually beat a grandmaster. And it was really shocking for everyone involved. Now, the interesting thing about what it did was during the game, it made a move that nobody expected. The grandmaster, the commentators, everyone, even the other players were utterly confused. That move just made absolutely no sense. And yet that was the move that won the game. And the point is, it (AI) can start doing things and making connections.


So, if you think about medicine as an example, it's (AI) got all the information in the world, it can start making connections between materials, diet, all these things and medical stuff that an individual just can't do because we're programmed in one domain.

“A few years ago, do you remember those Google deep minds? They had this Chinese game called “GO”. It's a very strategic game and very difficult to codify. And it actually beat a grandmaster. And it was really shocking for everyone involved. Now, the interesting thing about what it did was during the game, it made a move that nobody expected. The grandmaster, the commentators, everyone, even the other players were utterly confused. That move just made absolutely no sense. And yet that was the move that won the game. And the point is, it (AI) can start doing things and making connections.


So, if you think about medicine as an example, it's (AI) got all the information in the world, it can start making connections between materials, diet, all these things and medical stuff that an individual just can't do because we're programmed in one domain.

Check out what DeepMind, in collaboration with Google, achieved with GNoME development. They've made significant strides in crystal structure analysis, potentially revolutionising fields like semiconductor development and materials science. By leveraging AI models, they sifted through a staggering 2.2 million crystal structures, a feat equivalent to 800 years of traditional research effort. The speed and efficiency of this approach holds immense promise for accelerating advancements in various technologies and materials.


On another note, as AI continues to advance, we're witnessing its darker side in the realm of finance. Deep fakes, in particular, have garnered considerable attention due to their potential for fraudulent activities. However, it's not just the risk of fraudulent transactions; issues like plausible deniability also come to the forefront. Imagine someone disputing a payment authorisation, claiming it wasn't them. With deep fake technology, verifying identities and transactions becomes increasingly challenging, raising important questions about security and accountability.

Check out what DeepMind, in collaboration with Google, achieved with GNoME development. They've made significant strides in crystal structure analysis, potentially revolutionising fields like semiconductor development and materials science. By leveraging AI models, they sifted through a staggering 2.2 million crystal structures, a feat equivalent to 800 years of traditional research effort. The speed and efficiency of this approach holds immense promise for accelerating advancements in various technologies and materials.


On another note, as AI continues to advance, we're witnessing its darker side in the realm of finance. Deep fakes, in particular, have garnered considerable attention due to their potential for fraudulent activities. However, it's not just the risk of fraudulent transactions; issues like plausible deniability also come to the forefront. Imagine someone disputing a payment authorisation, claiming it wasn't them. With deep fake technology, verifying identities and transactions becomes increasingly challenging, raising important questions about security and accountability.

James Hobbs

HSBC

Vinay Parmar

Curo

Of a certain age, there's a distrust. If I speak to a machine, how do I know what’s going to be actioned?

Of a certain age, there's a distrust. If I speak to a machine, how do I know what’s going to be actioned?

I've got a similar point, and this is nothing to do with AI. Your point about a certain generation not trusting AI. There's also just healthy cynicism that things just don't work.


So I'm a season ticket holder with West Brom. I purchased a match ticket in recent times, and the system went automated, apparently automatically adding any purchase to your season ticket card automatically.


I remember when doing it thinking, that's not going to work - come the day and sure enough it didn't work.

Implementation is still key and you need a human as part of the process.

It has to be the right mix of automation and human implementation.

I've got a similar point, and this is nothing to do with AI. Your point about a certain generation not trusting AI. There's also just healthy cynicism that things just don't work.


So I'm a season ticket holder with West Brom. I purchased a match ticket in recent times, and the system went automated, apparently automatically adding any purchase to your season ticket card automatically.


I remember when doing it thinking, that's not going to work - come the day and sure enough it didn't work.

Implementation is still key and you need a human as part of the process.

It has to be the right mix of automation and human implementation.

Mike Lewis

WMGC

Greg Telford

Dr Sudheer Parwana

PwC

What I would say, though, is this is very impressive tech, and this is the worst it's ever going to be.

So, there's only one way it's going. And there's things like hallucinations, which we're all freaking out about at the moment. They're going to fix that because they know the big thing, stopping adoption is hallucinations. So these things will be sorted. And I think just standing still is simply non-negotiable.

What I would say, though, is this is very impressive tech, and this is the worst it's ever going to be.

So, there's only one way it's going. And there's things like hallucinations, which we're all freaking out about at the moment. They're going to fix that because they know the big thing, stopping adoption is hallucinations. So these things will be sorted. And I think just standing still is simply non-negotiable.

We then moved on to take part in a live poll...

What are the biggest opportunities for Enterprises in AI & Automation right now?

What are the biggest opportunities for Enterprises in AI & Automation right now?

Interesting seeing those results, I think that backs up what we’ve been talking about. That for now it is more around the automation and taking out that grunt work, whereas maybe the long-term opportunities, are more around research and development and moving these experimental techniques out of the back office and into the front office i.e. Customer facing.

Mark Simpson

GW

We then moved on to vote on the barriers...

What are the biggest barriers to blocking AI innovation in your business?

What are the biggest barriers to blocking AI innovation in your business?

Mark Simpson

GW

These results highlight the first difficulty is theory when it comes to AI use in business. The hard thing is applying it to your business problems, and very often you'll find data scientists will give you the data and application of it, but they won't use cases of how you can apply it. That's got to come from within your business, hasn't it?

The onus falls on businesses themselves to identify how AI can be leveraged to drive value and solve practical challenges within their operations. This educational process is essential, especially for industries like trading, where there may be a sense of apprehension or uncertainty surrounding AI adoption.

When introducing AI concepts to businesses, it's crucial to focus not just on the potential of data but also on practical use cases that demonstrate how AI can directly impact and improve business outcomes. By highlighting specific areas where AI can make a tangible difference, businesses can better grasp its relevance and potential benefits for their operations.

I think skill-set is hugely important here. The skill set piece is a big resource piece, isn't it? It's all very well looking for resources to inform your strategy and thinking, but the resources that are going to implement this design, are scarce and somewhat new.

We're changing our position with our engineering team now, we're not necessarily looking at experienced hires, but looking at either masters or PhD graduates in data science for those roles rather than experienced roles. They're going to come in at the same kind of benchmark price, but they're going to come in with some research knowledge in that field rather than looking for an experienced hire whose technology knowledge is probably going to be two years out of date compared to these graduates.

Jas Bassi

Gateley

Conclusion

Uniformly we all agreed that every business was a technology business and the lens that we should see AI through is, its in our toolbox. It’s a creative efficiency tool that when utilised with the right guardrails, big data, tech and people will enhance every offering.


What we are all searching for is those guardrails for our businesses and business cases to develop even further how we can increase efficiencies and effectiveness.


Some questions to think about...


  1. How is your IT department incorporating AI?

  2. What are your biggest opportunities?

  3. What are your concerns?


We’d love to hear your feedback.


#Birminghtechleaders

If you have any questions or queries please contact one of our Board Members:

Reuben Boughton

Reuben Boughton

Email Reuben
Mike Morgan

Mike Morgan

Email Mike

Stay in the loop

Join the conversation

Stay in the loop

Join the conversation

Powered by:

Got a question or want to get involved? Let us know!

Reuben Boughton

07970 254394

reuben.boughton@griffiths-waite.co.uk

Mike Morgan

07852907506

mike.morgan@hays.co.uk

Reuben Boughton

07970 254394

reuben.boughton@griffiths-waite.co.uk

Mike Morgan

07852907506

mike.morgan@hays.co.uk

Got a question or want to get involved? Let us know!

Powered by:

Back to top

Reuben Boughton

07970 254394

reuben.boughton@griffiths-waite.co.uk

Mike Morgan

07852907506

mike.morgan@hays.co.uk

Got a question or want to get involved? Let us know!

Powered by:

Back to top