AI
Problem Lenses: How and when should I use AI? - Part 3
Part 3 of Ali Salaman’s three-part series looks at how businesses can use AI to explore ideas and manage knowledge with large language models
Welcome to the final part of our blog series on how to assess if large language models (LLMs) like ChatGPT can solve your business challenges. In Part 1 we explored lenses focused on automating repetitive tasks, supporting decision-making, analysing unstructured data; in Part 2 we looked at optimising workflows, mitigating risks, and breaking down language barriers.
In this post, we’ll look at two final lenses: The Lens of Thought Exploration and The Lens of Knowledge Management.
These lenses showcase how LLMs can be a catalyst for creative thinking and a valuable tool for managing information effectively. Let’s explore how you can apply these ideas in practical ways.
The Lens of Thought Exploration
What is this lens?
This lens focuses on using LLMs to explore ideas, brainstorm strategies, and simulate different scenarios. LLMs can help you generate creative solutions to complex problems, consider “what-if” scenarios, or refine your plans by identifying overlooked perspectives. Unlike human colleagues, they are objective, so do not hold opinions or develop personal attachments to ideas, providing a safe environment to explore potentially effective solutions.
Practical steps to apply this lens:
1. Brainstorming ideas: Leverage the power of LLMs to generate creative ideas by prompting it with open-ended questions. For example, a council aiming to increase recycling rates might ask: “What are three ways councils in the UK could encourage households to recycle more?”.
Prompt templates:
2. Simulate Scenarios: Use the LLM to explore and evaluate “what-if” situations. This is another strength of LLMs and they can be used to provide insights into the possible outcomes of decisions you may be considering. For example, a council considering pedestrianisation of a high street in a market town could ask: “What are the economic implications of pedestrianising the high street?”
Prompt templates:
3. Refine Strategies: Use an LLM to suggest improvements or alternatives to existing plans, saving time and avoiding the biases or sensitivities that can arise in brainstorming sessions. For instance, a small business expanding its e-commerce operations in Scotland and Northern Ireland might ask: “How could our online strategy better cater to customers in these regions?” The LLM could propose data-driven strategies such as optimising delivery networks by considering population density and geographic challenges, targeting localised advertising by analysing regional search trends, or recommending regionally popular products based on purchasing patterns. This allows the business to quickly surface actionable insights that might otherwise require extensive research or group workshops to uncover.
Prompt templates:
4. Facilitate Cross-Functional Collaboration: Use LLMs as a neutral party to generate ideas that bridge gaps between teams with differing priorities. You may have two teams in different departments with different or opposing views about the direction of a particular project or struggling to buy in to the objective. You can explore ways that you can use to help them to work together more effectively or to see your vision and bring them onboard.
Prompt templates:
5. Capture Insights Quickly: During brainstorming sessions, use LLMs to document and organise ideas in real-time. After a discussion on improving NHS staff wellbeing, a prompt like “Summarise the main challenges raised and suggest three priority initiatives to address burnout” could yield a concise summary, ensuring no key points are lost and enabling immediate follow-up actions.
Prompt templates:
Examples:
Policy Development: A local council could use LLMs to generate policy ideas for enhancing rural public transport. Asking “What initiatives could improve access to public transport in remote areas?” might result in proposals such as subsidised tickets for low-income residents, on-demand minibuses tailored to rural communities, or partnerships with local taxi services.
Business Strategy: A retail company could brainstorm ideas for expanding into new markets by asking an LLM to suggest entry strategies and potential challenges.
Using LLMs for thought exploration empowers your team to think creatively without being hindered by apprehension for fear of looking stupid, explore diverse perspectives, and make more informed decisions.
The Lens of Knowledge Management
What is this lens?
This lens focuses on how LLMs can help your organisation organise, retrieve, and utilise knowledge more effectively. Whether it’s summarising lengthy reports, tagging documents, or answering complex questions based on company data, LLMs can transform how your team accesses information.
Practical steps to apply this lens:
Streamline document summarisation: Use LLMs to summarise large volumes of reports, articles, or meeting notes into concise summaries. For example, a prompt such as “Summarise this 50-page financial report into a one-page executive summary” could save hours of manual work and ensure stakeholders quickly grasp the essentials.
Enable knowledge retrieval: Integrate LLMs with internal knowledge bases to allow employees to ask natural language questions and get accurate answers. Implemented correctly and with the right solution, this can transform the way your organisation and even your customers can access knowledge for day to day needs.
Tag and categorise information: Automate the organisation of unstructured data by tagging documents with relevant keywords or categories. User enquiries can be categorised with topics like “Onboarding Issue”, “Technical Issue,” or “Account Setup” for faster triage. Subsequent knowledge of this information can then lead to a roadmap of work to address commonly occurring issues.
Provide contextual answers: Providing extracts of company information in a prompt sent to the LLM can facilitate more contextually aware responses. This could be information in company documents or a user record in a database. Of course data privacy can be a concern here, but when using an LLM deployed in your own environment that you know will not leak data to third parties, you can free-up your organisation to utilise the capabilities of LLMs.
Facilitate team onboarding: New employees can use LLMs to quickly learn about company policies, processes, and resources by asking questions in plain language. For example, asking “How do I submit a travel expense reimbursement claim?”
Example Applications
Internal knowledge bases: A healthcare organisation could use LLMs to help staff quickly access clinical guidelines or research papers.
Customer support: A tech company could use LLMs to analyse past support tickets and create a comprehensive FAQ for both customers and employees.
By streamlining knowledge management, LLMs can save your team time, reduce information silos, and ensure faster, more accurate decision-making. Solutions like Librian, the new AI solution from Caution Your Blast Ltd (CYB), can unlock safe, reliable access to your knowledge bases to make more effective use of it within your organisation and for your customers.
Conclusion
Throughout this three-part blog series, we've delved into how artificial intelligence (AI) can be a transformative tool for businesses looking to innovate, streamline operations, and make more informed decisions. By examining your challenges through the various "problem lenses" I have outlined, you can identify where AI might offer tangible benefits.
The journey to integrating AI into your organisation starts with a thoughtful assessment of your current processes. Begin by mapping out your core activities and pinpointing areas where AI could make a significant impact. Remember to validate that these processes are effective in the first place; automating a flawed process won't yield better results. Start small with low-risk pilot projects to test AI's capabilities within your specific context, allowing you to build confidence and learn valuable lessons before scaling up.
Ethical considerations should remain at the forefront of any AI implementation. Maintain human oversight, ensure data privacy, and be transparent about how AI is used within your organisation. By doing so, you not only mitigate risks but also build trust among your stakeholders.
Embracing AI isn't just about adopting new technology - it's about empowering your business to work smarter, innovate faster, and serve your customers better. We encourage you to apply the insights from this series to explore how AI can address your unique challenges. If you're ready to take the next step, consider reaching out to experts like Caution Your Blast for tailored guidance on implementing AI responsibly and effectively. Let’s unlock the potential of AI together and drive your business forward in the digital age!
Caution Your Blast Ltd's Librian AI Solution can revolutionise your customer service by providing fast, accurate, safe and ethical responses to user questions - it is proven to reduce in-bound written enquiries by 80%. Request a demo or get in touch here