Zeeshan Anwar
Chief Executive Officer at NexaQuanta | AI Technology Leader
Insights on Generative AI: An Introduction
If you were to take my LinkedIn profile at face value, you’d think I’ve been committed to AI for more than two decades. And in part, you’d be right – after all, I did do my Masters in Engineering and Artificial Intelligence. Scratch the surface though, and you’ll see that the reason I’m so invested in AI is its ability to solve human problems. From fixing the issue of mobile devices overheating, right through to helping the compliance and audit professionals in automating the complex and mundane tasks, I am forever excited by the opportunity to make people’s lives easier with the help of advanced AI technologies.
And while it is normal for some to fear that Generative AI (commonly referred to as GenAI) will replace jobs – I see enormous potential for it to enhance our working lives. Because, for all of our technological advancements, people and organisations still face barriers to productivity and experience friction that makes it difficult for them to perform to their best ability. This in turn leads to inefficiencies and risk factors that negatively impact employee satisfaction and erode the bottom line.
57% of employees believe the difficulty of finding the correct information is one of the biggest contributors to lagging productivity
What can GenAI do – and where can organisations start?
From a human angle, we know people want the ability to question data using natural language. Recent research shows that nearly a third of UK employees are sharing private company data with ChatGPT and other publicly available online GenAI solutions. But before running off to see if that’s a problem for your organisation, the key thing to understand is why an employee would use a public-facing Large Language Model (LLM) in the first place.
Putting the novelty of new technology to one side, what we see is employees who are trying to solve problems caused by a lack of:
- access to the data they need
- ability to interpret the data in a way that makes sense
- time (and appetite) to find and filter data
- knowledge of data location and reliability
With data being generated at a faster rate than any human could manage (and stored in too many places to keep track of), it’s no surprise that people turn to shadow IT.
The good news is that GenAI can help to solve these kinds of problems. The challenge is that many organisations feel unsure of how to go from idea to making GenAI a reality. So, here are three quick steps designed to help put you on the right path:
1: Start small
My advice to organisations is to begin with a simple, low-cost use case, as this provides the fastest return-to-value at the lowest risk. It also helpfully moves the conversation away from the worry of AI replacing jobs – the use cases focus on making life simpler.
Examples include:
- Connecting employees with your knowledge base and letting them ask questions and have easy access to the information they need
- Automating simple and mundane tasks of data collection and processing such as summarising and translating documents, creating templates for standard forms such as emails and proposals
2: Choose enterprise-ready
It is natural for enterprises to be wary of relying on one Large Language Model, or using one which has been trained in the public domain. We choose to use IBM watsonx for its enterprise-ready capabilities, which include (but are not limited to) the following:
- Uses multiple LLMs with different areas of specialisation (ie: following instructions, writing code, asking questions)
- IBM Granite LLM has been trained on enterprise data to do enterprise jobs
- IBM offers full indemnity – the only vendor to so
- IBM watsonx.governance provides the ability to direct, manage and monitor AI activities wherever, and however, they are deployed
- Compatible with all hyperscalers and integrates with key platforms and apps such as Microsoft Teams and Slack
3: Have a human conversation!
As Anne Leslie CISM CCSP, Cloud Risk & Controls Leader, EMEA at IBM said on a recent call, “What lots of organisations need to be successful with GenAI are better questions and human conversations”. It’s just one of the reasons that I invited her to join me for a special on-demand webinar and IBM watsonx AI Assistant demonstration.
During the 30-minute session, we use the forthcoming EU DORA regulation as a use-case for what IBM watsonx AI Assistant and knowledge-based access can do. My thanks go to Anne for her clarity in explaining the regulation, what it means in the overall context of operational resilience and how it can act as a catalyst for those important conversations. There is also a demo from me of the NexaQuanta solution built on watsonx AI Assistant for safe and secure access to the documents in your knowledge base to highlight the capabilities of GenAI with IBM watsonx AI Assistant.
To watch the webinar and demo at a time that works for you, simply follow this link https://nexaquanta.inbound.systems/#webinar – and if you want to follow that up with a conversation on the possibilities of GenAI with IBM watsonx AI Assistant – drop me a DM, I’d love to talk!
