AI Unleashed: Transforming GBS With Generative AI

The amount of information on Generative AI has reached a tipping point. That is, there is now so much, you need a bot to be able to read it all, sift out the facts, and provide some useful context. That's why clear, practical, and useful advice on how Generative AI can be applied specifically to Global Business Services, is so vital.

In order to gain more knowledge on the opportunities and potential risks Generative AI creates for GBS, the SSONext podcast welcomed Phil Searle, CEO and Founder of Chazey Partners.  You can listen to the podcast below, but read on for our biggest takeaways from the episode. 

Immediate Opportunities Gen AI Offers

As organizations look to make the most out of their new AI technologies, it’s important to understand how Generative AI fits into the wider tech solution space. For Phil Searle, there are three types of technologies.

1. Traditional ERP – Solutions such as SAP and Oracle represent traditional ERP technologies. These solutions are typically installed on-premise and are managed by businesses IT function.
2. Mid-range technology – Solutions that provide services such as case management, workflow automation and document management. These solutions are similar to the third type of technology but not as advanced.
3. Intelligent automation - This category refers to the latest business solutions technologies. These include solutions such as RPA, Machine Learning, and Intelligent Document Processing (IDP).

According to Phil, Generative AI is offering GBS the opportunity to take their intelligent automation abilities to the next level, by bridging the value gap between what the promise of IA and the reality:

“Optical Character Recognition (OCR) has been great, but it has never been completely seamless,” said Phil. “Now, with Generative AI combining OCR, Natural Language Processing (NLP), and computer vision, you can improve the success rate of data capturing.”

As a result of improved data-collecting abilities, GBS can improve its data analysis capabilities and ultimately strengthen its decision-making.

Another place GBS can see immediate dividends with Generative AI is in the customer service space. Organizations have the chance to improve chatbots so they can understand context, tone, and much more about the customer they are engaging with. 

The third area Phil mentioned that GBS can implement Generative AI is in supply chain. Generative AI can be used as a forecasting tool so businesses can predict and navigate any challenges they might face while overseeing the supply chain or any other logistical areas.

“Some companies, for example, are using generative AI to generate algorithms for demand forecasting, waste reduction, inventory management, transportation management and actually reducing cost and emissions of transportation networks,” Phil said. 

Expected Challenges of Working With AI

When asked about risk factors tied to Generative AI, Phil offered the following list of his biggest concerns:

• Copyright and patent infringement: Generative AI models are trained on large datasets of text and code. This means that they are capable of generating content that is similar to or even identical to copyrighted or patented material. This could lead to legal problems for businesses that use generative AI without permission.
• False certainty: Generative AI models can be very convincing, even when they are generating inaccurate or misleading information. This is because they are trained on large datasets of text and code, which can include both accurate and inaccurate information. It is important to be critical of the output of generative AI models and to verify the information they generate before using it.
• Hallucination: Generative AI models can sometimes generate content that is completely fabricated. This is called "hallucination." Hallucination can be a problem if it is used to create fake news or other forms of misinformation.
• Bias: Generative AI models are trained on data that is created by humans. This means that they can inherit the biases that are present in the data. For example, a generative AI model that is trained on a dataset of news articles may be more likely to generate text that is biased towards a particular political party or ideology.

If you want to hear Phil explain more of the risk factors, deployment opportunities, and best fits for Generative AI, be sure to check out the entire episode.