AI is a support for businesses but not a solution
AI is a useful support for businesses but not a quick-fix solution, say leading tax and advisory firm Blick Rothenberg.
Nick Winters, head of technology at the firm, said: “Microsoft has unveiled a new Artificial Intelligence (AI) hub in London this month to take advantage of the UK’s vast AI talent and expertise. But whilst the development of AI in the UK is exciting, when it goes wrong, it can cause more problems than it solves.
“Currently, AI is no substitute for the human brain and needs to be used in context; it should not be used as a stand-alone tool but in conjunction with traditional thought processes.”
He added: “I have some personal examples of when AI used on its own has caused major problems. One of my clients informed me that they could use a well-known accounting software to prepare and submit their tax return. They later called me in a panic when their tax liability was a multiple higher than expected. A brief review revealed that the software had identified all bank receipts as income, even those that were just a transfer from savings to current account.
“I sometimes use Bing’s AI companion to help me pull together ideas for articles or help sheets. I asked it to provide something that might be used as the financial policies for a party manifesto. At first glance, what I received seemed interesting, but on closer examination, it was Kwasi Kwarteng’s 2022 mini-budget, and we all know how that was received.
“There is also the example of New York lawyers who were sanctioned after using the AI tool ChatGPT for legal research. The filing was found to reference examples of legal cases that did not exist!”
Nick said: “However, we have all already benefitted from the support offered by AI technology; everyday examples include– satellite navigation systems to driver-assisted technology, in the home – robot vacuums and smart thermostats, opening your phone with face ID, enhanced internet search capabilities and in the workplace – spellcheckers, predictive text and chatbots, to name but a few.”
“So, when used properly, AI can help us become more productive and live a better life – but only if it is used as a support, not a quick fix to our problems.”
He added: “Therefore, my top ten tips for businesses using AI or who want to start using it are:
- Define Clear Objectives: Identify specific problems or opportunities where AI can add value to your business, whether it’s improving customer service, streamlining operations, or enhancing product recommendations. Don’t implement AI technology without checking if it is an appropriate support in the first place.
- Data Quality and Accessibility: Ensure your data is clean, relevant, and accessible for AI algorithms. Invest in data collection, storage, and management systems to support AI initiatives. This can help to avoid model drift, where poor quality or irrelevant data leads to an AI giving incorrect or inappropriate responses.
- Start Small, Scale Gradually: Begin with pilot projects or small-scale implementations to test the feasibility and impact of AI applications. Once successful, gradually scale up AI initiatives across different departments or business functions.
- Choose the Right AI Tools and Partners: Select AI tools, platforms, or vendors that align with your business needs, budget, and technical capabilities. Consider factors such as ease of integration, scalability, and vendor support.
- Invest in Talent and Training: Build internal expertise in AI by hiring data scientists, machine learning engineers, and AI specialists, depending on your budget. Provide training programs for existing employees to upskill and adapt to AI technologies.
- Ethical and Responsible AI Use: Prioritize ethical considerations and responsible AI practices, including data privacy, transparency, fairness, and bias mitigation. Ensure compliance with relevant regulations and standards. Human intervention is still needed to moderate AI output before it is published to fact-check and ensure it is appropriate.
- Iterate and Learn: Embrace a culture of continuous learning and iteration in AI initiatives. Encourage experimentation, feedback, and collaboration across teams to refine AI models and strategies over time.
- Monitor Performance and ROI: Establish metrics and KPIs to measure the performance and ROI of AI implementations. Regularly monitor and analyse data to assess the effectiveness and impact of AI initiatives on business outcomes.
- Stay Informed and Adapt: Keep abreast of the latest developments and trends in AI technologies, applications, and best practices. Continuously evaluate and adapt your AI strategy to stay competitive and address evolving business needs.
- Customer-Centric Approach: Keep the customer at the centre of your AI strategy. Focus on delivering personalised experiences and anticipating needs. As before, use AI as a support, not a solution – customers’ issues with your business cannot be fixed with an AI band-aid. ”
Nick said: “By keeping in mind that AI is a support, not a solution, businesses can harness the power of AI to drive innovation, efficiency, and growth while avoiding the potential pitfalls.”
