Over the past few years, my team has been on a journey to develop and scale a multi-model, multi-modal Generative AI framework for business, aiming to accelerate productivity and unlock insights from data. This includes Agentic AI capabilities, where AI “agents” can autonomously take action and learn.

Although the technology is great, it is only valuable when applied to support a specific outcome.

Even with the mainstream media hype, it is still evident that many individuals struggle to identify how Generative AI might help them personally. Therefore, I thought I would share my common use cases, many of which I believe are applicable to a wider audience.

Summarisation, Comparison and Analysis

  • Reference unstructured documents (e.g. Word, Excel, PowerPoint, PDF) to summarise, compare and/or analyse the content. Support the consumption of information, helping to highlight key messages, themes and trends. Examples include white papers, blogs, contracts, policies, earnings, etc.

Writing

  • Reference existing sources (e.g., documents, communications, interviews) to draft and/or critically appraise written material, incorporating personal style and inflections. This includes sentiment analysis of draft material to help improve/refine the messaging for specific audiences. Examples include communications, blog posts, policies, etc.

Research and Reasoning

  • Reference private and/or public data sources to prepare a detailed research document regarding a specific topic. This could include a concept (e.g., Quantum Computing), company, product or service. Context can be provided to tighten the scope of the research, identifying strengths, weaknesses, opportunities, threats, etc.

Structured Data Insights

  • Connect to structured data sources, providing a natural language interface to query and analyse data, including the dependencies and/or connections between data sets.

Knowledge Base

  • Reference a specific knowledge repository to create an expert system focused on a specific topic. Valuable when interrogating large knowledge articles, manuals, readme documentation, processes, procedures, etc.

Coding Assistant

  • A software development assistant, specialising in a specific programming/scripting language. Supporting code generation, suggestions, reviews, testing, debugging and documentation creation.

Ideation and Problem Solving

  • A mechanism to support brainstorming, quickly generating hundreds of ideas focused on a specific problem or outcome.

Image and Video Generation

  • The creation of images and/or videos for use within presentations, blogs, etc. Side note, the header image for this article was AI generated.

At this point, I can confidently state the use of Generative AI acts as a “force multiplier”. I am able to deliver more, with less effort. It also reduces pressure by automating low-value, but often time-consuming tasks, whilst also frequently improving the quality of the output as I have more time to apply critical thinking.

It should be noted, as part of these use cases, that Generative AI does not replace my individual expertise, experience or style. Instead, it augments my thinking, combining the best of human and machine intelligence.

Although I leverage our custom-built capabilities, with access to advanced AI models, the majority of these use cases can be achieved with consumer capabilities, offered by OpenAI ChatGPT, Google Gemini, Microsoft Copilot, etc.

enterprise architecture ea enterprise_architecture artificial_intelligence ai machine_learning ml generative_ai agentic_ai large_language_model llm natural_language_processing nlp transformer data prompt_engineering openai chatgpt gpt o1 microsoft copilot google gemini gemini_pro imagen imagen_2 imagen_3 ai_studio project_mariner bard lifeinwork