Assessing your organization's AI Readiness
Assessing your organization's AI Readiness
When chatGPT hit the news in November 2022 everyone was rushing to try it out and integrate it to their routine workflow. I remained skeptical for about four months before I wrote my first prompt. Even then I was still using it like a normal search engine. At that time the AI train had taken off and panic was already in the air that jobs were already at risk of being replaced by AI. For a tech professional it was quite absurd that I didn’t immediately jump into the panic that I was about to lose my job to AI. I simply wasn’t convinced that a ‘chatbot’ as I thought of chatGPT at the time was capable of taking my job. Well it cannot even open an IDE or collect requirements from end users, so I thought. More than three years later I am still convinced that humans aren’t completely replaceable. There are opportunities for me and many other professionals to responsibly and intelligently integrate AI into our work for efficiency. AI agents are the new shiny tools in the AI space. I can implement an agent that completes end to end operations that a recruitment associate would complete;
- Going through hundreds of resumes
- Identifying the candidates that are most relevant to the job
- Scheduling interviews
- Sending email updates to candidates
It is very important for organizations to consider their preparedness and relevance to business operations when implementing AI in their workflows. Consider a situation where a company becomes convinced that they can utilize a coding agent to automate their programming tasks instead of hiring software developers. In the background the organization does not have its codebase organized in a repository and lots of sensitive configuration data is hardcoded. Developers have simply had to patch things just to keep systems running. Exposing such a codebase to a coding agent will be messy and runs the risk of exposing sensitive data to publicly available AI models that are in continuous training. This example forms the basis of how and why every organization should evaluate its AI readiness even before the automation process begins.
AI tools are an expensive investment that needs proper planning to avoid wastage of resources. AI preparedness for any organization involves looking at aspects such as data availability and quality, data governance procedures, ethical considerations, alignment of AI tools to business operations and team technical capabilities. AI tools need to be trained on quality data to achieve correct outputs. An organization cannot rely on responses from an AI tool that has not been exposed to its data and business scenarios. Some lines of business require a very careful approach on AI implementation because overreliance on AI might lose client trust. In other cases end to end automation using AI tools is possible and a welcome reprieve to the business. In the process of implementing AI the organization might discover that some of the roles are no longer necessary as the tasks can efficiently be completed by AI. It requires a proper change management procedure to help employees understand that retrenchment is not based on their performance or skills but rather on changing business needs. The remaining staff would ultimately start questioning their role in the organization if routine tasks are completely overtaken by AI.
AI Readiness the Right Way
In some instances stakeholders have found themselves taking a reactionary approach to the trajectory of AI utilization within their team’s workflows. Without a proper formalized guideline on how to utilize AI, sometimes employees find themselves just tinkering around with tools and flowing with whatever captivates their vibe. By the time stakeholders realize that each employee has their own understanding and approach to how they utilize AI then it is already too late to come up with the organizational framework on AI usage. For jobs that require expertise to be built over time stakeholders are suggesting a supervised approach on how junior employees utilize AI in their tasks. Other organizations have completely suspended AI usage for client work, with tough repercussions on any AI generated content. AI knowledge is everywhere and accessible to everyone. Therefore,suspending AI usage the right way might not be the right approach as it limits innovation.
What is our AI strategy? Are we the organization that has specific tools that all employees should use or do we give freedom for each individual to explore the diverse AI tools available. Does our AI strategy define what tasks can or cannot be completed using AI? Do we have a defined code of ethics and repercussions when these rules are not followed? What is the status of our data? Do we have available and quality data that can be leveraged to train AI tools that are usable in our organization’s use cases. Have we defined data governance policies that define who and how the data should be accessed? Is our team properly trained and equipped to handle AI Tools? Just like any other new tool, teams require proper training and capacity building to correctly handle AI. Have all team members demonstrated personal responsibility and ability to implement AI in an ethical manner. What is the organization’s technical capability to handle AI? In addition to just using off the shelf AI tools the organization needs a qualified team to properly integrate and scale AI tools within its business workflow.
Conclusion
AI is transformative and each organization needs to ensure it is well equipped to implement AI the right way. Limiting AI usage or prohibiting it entirely will only lock the organizations from attaining the benefits that would be gained in automating its workflow. Strategy, data quality, team capacity, and technical capability are some of the important aspects that an organization needs to evaluate in its AI readiness assessment.