Businesses have seen a steady increase in requests to implement Gen AI. However, right out of the gate they face a stumbling block: What do we use this for? Though many have a vague idea, the question around the core value prop comes a bit late. There are a few fundamental questions which businesses should ask which we have tried putting together.
-What does Generative AI mean to me?
Understanding how Generative AI works and how it impacts work within one’s own paradigm is essential. At the most fundamental level, generative AI leverages neural networks to produce original content in the form of text, videos, music, speech and even software code. Moreover, Generative AI heavily depends on the training data you provide it. Which is where your personal understanding of Gen AI is essential for you to answer the question “Generative AI can help write, create, answer ____ for me which usually takes me or my team 100 hours to do”.
-Where do I want to be?
Imagine a world where your company has already reached its zenith of value creation. In this scenario, what is Generative AI doing for you? Those can be small jobs to very high value creation for the company. This fanciful future for your company will help build a special pathway to your company using the following steps:
Keep in mind, this is an oversimplification of the steps one might follow, but is rough method you would be on.
-What problem should I pick?
This tends to be both the easiest and hardest steps. Off the cuff ideas are aplenty but we end up overthinking this step. The minimum criteria for picking the first problem should be:
If you reach a YNY, then the problem is a good one to start with. Importance of the problem might have been a good parameter, but it is important to get your feet wet with Gen AI before changing its focus towards more core problem areas. Gain your experience with your lower hanging fruits. Picking the perfect problem is a separate set of evaluations which you can get from the maturity index mentioned in the previous point.
-How do I start?
Starting is easier than it sounds. Get an expert! If your team has the technical know-how, then they can train your specialised bots on OpenAI or Llama 2 and a dozen other AI LLMs. But experts will help you pick the right dataset, help you with understanding terminologies, and involve you as the collaborator with AI rather doing everything yourself.
I realise its easy to ask you to get an expert to kick things off before you have thought of the first two questions. However, there is a mental maturity in accepting Gen AI which is needed which requires a thoughtful approach. Moreover, Gen AI doesn’t always give you the best result, or sometimes hallucinates its outcomes which needs some forewarning and diligence from the team. All these come together well, once you have thought through your approach and then engage with expert teams to build contextualised and specialised Gen AI solutions. Ideally your POCs should not take more than 2-3 months to implement, though there is some wiggle room based on complexities of execution and data integrations required. Simpler the better!
Conclusion: Gen AI is a transformative technology but it’s only useful if you know what you are transforming. A strategic and thoughtful approach to Gen AI is essential for your company to build a long term effective strategy to create impact.
The magic in the name will last only till you start realising gains from using it. So knowing those gains is essential. Diagnosing where and how you get there, is something we can help with.
Read more insights from network science teams of experts.
