Governing GenAI: Risks and opportunities for SMEs
Running a small or medium-sized enterprise in today's fast-paced world is no small feat. You're constantly looking for ways to innovate, stay competitive and meet the ever-evolving needs of your customers. Generative AI (GenAI) offers a powerful solution, but it also presents new challenges. As a result, company directors are finding it raises critical questions:
- How can it best serve their business?
- How is it being used effectively?
- What are the risks, and how can they ensure proper oversight when expertise is scarce?
Recent Diligent research among public company directors found that:
Those figures – all below the 50% mark – show that a lot of companies are still in the starting gate when it comes to AI.
Diligent, in partnership with the Institute of Directors, recently convened a panel of experts and invited its members to join them for a confidential discussion about the challenges and opportunities around AI in business. For an in-depth look at the insights as well as practical strategies from the discussion, download our AI Unleashed: Tips for small and medium enterprises report here.
In advance of the debate, the panel set the scene by illustrating the reality of AI in businesses today. Key takeaways included:
AI is not new… and it’s not just GenAI
One expert reminded attendees that AI is not new, it’s been on a 70-year journey that incorporates many technologies we use daily, such as machine learning and algorithms. “It’s not one thing and therein lies the problem for businesses in understanding what they are buying and using,” they said.
The OECD’s definition of AI provides clarification: “An AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. Different AI systems vary in their levels of autonomy and adaptiveness after deployment.”
This definition is now enshrined in EU law, which ensures everyone is talking about AI from the same starting point. The recent emergence of commercially viable functional AI, which can act autonomously and create something “new” in response to prompts, has elevated awareness and created the need for stronger governance and a more urgent need for businesses to develop clear knowledge about AI and its safe application.
AI-driven growth for your business
Based on a survey of 60 directors, this report offers expert insights on responsibly leveraging AI, covering data protection, upskilling, transparency, and governance.
Download your copy todayCombine traditional AI uses and new GenAI capabilities to find the sweet spot
Building on the introduction, another expert charted the evolution of AI in business as the “cornerstone of digital transformation”. It has been used to optimise operations, reduce costs and make data-driven decisions in applications such as demand forecasting, stock management and dynamic pricing, alongside quality control and predictive maintenance. These are largely business-facing use cases focused on efficiency and productivity.
GenAI introduces the opportunity for deployment in customer-facing use cases through its ability to mimic human creativity. Examples include highly personalised customer experiences and virtual assistants that advise customers based on an intimate understanding of their preferences and buying habits. “GenAI excels in creativity and personalisation,” they said, “helping businesses to adapt to customer needs much quicker while shaping and driving deeper engagements and ultimately competitive advantage.” They believe:
Start with low-risk, high-reward projects
Both traditional and generative AI offer tempting transformative potential, but this comes with risks and rewards. The advice is that companies don’t need to make everything AI-powered all at once. Instead, they should take a strategic approach: “Look at the problem you are trying to solve first; don’t just say, ‘Here’s a technology, where does it fit?’”. The suggestion is to select “low risk, high reward projects that allow the organisation to achieve confidence and competence.”
These early projects should also act as pilots for robust governance and transparency, enabling the organisation to establish the right level of oversight so the power of AI can be harnessed safely.
Balance enthusiasm with caution
Another expert shared some of the cautionary tales that have already emerged from corporate GenAI use, saying: “The real danger is of GenAI being thought of as human intelligence because it just isn’t. But its apparent fluency of text or audio answer conceals the nature of the statistical model behind it that can err like any software, and it happens quite painfully when it comes to light.”
They spoke of their personal experience, using AI-powered translation to translate 3000 hours of highly technical lectures into 22 languages in a very short time at low cost. However, they acknowledged that there is still a significant quality control exercise to undertake to check that the translations are accurate.
There are notable cases where GenAI is outperforming human counterparts. A customer service chatbot is now gaining higher customer satisfaction scores than human agents. However, caution was advised that companies are legally responsible when chatbots make mistakes – a defence from a company that it wasn’t responsible for a decision made by a chatbot was rejected by the court.
Ensure there’s a human in the loop
Summing up a common theme across GenAI use cases, another expert noted the need for a human in the loop to oversee, challenge and govern content and decisions made by AI. They raised the implications for training and development: “We need to look at AI literacy: do humans know how it is working?”
Employees require a level of understanding of the capabilities and limitations of AI systems so they can spot when errors have occurred, understand why they happened, and implement fixes where needed. AI implementation strategies should therefore include training and awareness programmes for staff who will be working with AI.
In a final thought, another expert recalled a recent comment they heard on the impact of GenAI:
Ensure your organisation stays ahead by complying with the EU AI Act, the first comprehensive regulation to address AI risks and safeguard fundamental rights within the EU. Find out how Diligent can help here.
Keep exploring
Governing AI: Obligations, ownership and oversight
Discover key takeaways from our panel discussion on introducing AI tools into your business and establishing effective AI governance.
Incorporating gen AI into business strategy
In this episode of The Corporate Director Podcast, we talk with Sophia Velastegui, a renowned advisor on AI business strategy with experiences at Microsoft, Google, and Apple.
AI in governance: Opportunities and challenges for modern boards
Learn how AI tools can optimise board performance, reduce risks, and streamline meeting documentation.
