What AI actually costs in operations
Implementing AI in operations often incurs significant costs in data preparation and integration, which can be mitigated by carefully planning and executing the deployment, for instance, a logistics manager can use AsscherAi to identify bottlenecks in the supply chain and optimise routes, reducing costs and improving efficiency.
In theory, AI systems like AsscherAi are supposed to streamline operations and reduce costs, but in practice, the actual cost of implementing and maintaining these systems can be staggering, with spend accumulating in unexpected areas, and most people are wrong to think that the cost of the AI model itself is the biggest expense.
As someone who has sat through numerous meetings discussing the deployment of AI in operations, I can attest that the majority of the cost is actually tied up in integration and data work, which can be a major hurdle for organisations looking to implement AI solutions, and this is where the real cost of AI in operations lies, not in the model itself, but in the work required to get it up and running.
Integration and data work as the dominant line
The cost of integrating AI systems with existing infrastructure and processes can be significant, and this is often overlooked when organisations are calculating the cost of implementing AI, but it is a crucial aspect of the overall cost, as it requires significant resources and expertise to ensure that the AI system is properly connected to the rest of the organisation's systems and data sources.
For example, a shift supervisor at a manufacturing plant may need to check the maintenance log to ensure that the equipment is properly maintained, but if the AI system is not properly integrated with the maintenance log, then the supervisor will not be able to get the information they need, and this can lead to delays and inefficiencies, which can increase the overall cost of the AI system.
Run cost against query volume
Another area where costs can accumulate is in the run cost of the AI system, which is often tied to the volume of queries that the system receives, and this can be a significant expense, especially if the system is handling a large number of queries, and organisations need to carefully consider the cost of running the AI system against the volume of queries it receives, to ensure that they are getting the best value for their money.
A category manager at a retail company may compare two suppliers to determine which one is offering the best price for a particular product, but if the AI system is not properly configured to handle the volume of queries, then it may not be able to provide the manager with the information they need in a timely manner, and this can lead to missed opportunities and increased costs.
The cost of a wrong answer nobody caught
One of the most significant costs of AI in operations is the cost of a wrong answer that nobody caught, which can have serious consequences for an organisation, and this is an area where most people are wrong to think that the cost is minimal, but in reality, it can be very significant, and organisations need to have processes in place to ensure that the AI system is providing accurate and reliable information.
For instance, if an AI system provides a wrong answer to a query, and nobody catches the mistake, then it can lead to incorrect decisions being made, which can have serious consequences, and this is why it is so important for organisations to have robust processes in place to ensure the accuracy and reliability of the AI system, and to visit our operations page to learn more about how to implement AI in operations.
Attributing spend to a process rather than a model
When it comes to attributing spend to a process rather than a model, organisations need to carefully consider the costs associated with each process, and ensure that they are properly accounting for all of the expenses, and this can be a complex task, especially in large organisations with many different processes and systems.
By taking a careful and detailed approach to attributing spend to a process rather than a model, organisations can get a better understanding of where their money is being spent, and make more informed decisions about how to allocate their resources, and this is an area where AsscherAi can provide significant value, by helping organisations to understand their data and make better decisions.
Organisations may use a variety of methods to attribute spend to a process, including
- activity-based costing
- process mapping
- cost accounting
When AI is the wrong choice
There are certain situations where AI is not the best solution, and organisations need to carefully consider whether AI is the right choice for their particular needs, and this is an area where most people are wrong to think that AI is always the answer, but in reality, there are many situations where other solutions may be more appropriate.
For example, if an organisation has a simple and well-defined process, then it may not be necessary to use AI, and a more traditional approach may be more effective, and in these situations, organisations should carefully consider their options and choose the solution that best meets their needs, and to contact us at contact-us to learn more about how to determine whether AI is the right choice for their organisation.
Frequently asked questions
What are the main costs associated with implementing AI in operations?
The main costs are integration and data work, which can be significant and often overlooked, requiring substantial resources and expertise.
How can organisations reduce the cost of running an AI system?
Organisations can reduce costs by carefully considering the volume of queries and configuring the system to handle them efficiently, ensuring they get the best value for their money.
What is the most significant cost of AI in operations that is often underestimated?
The cost of a wrong answer that nobody caught is a significant cost, which can have serious consequences, and organisations need to have processes in place to ensure accuracy and reliability.
When is AI not the best solution for an organisation's operations?
AI is not the best solution when an organisation has a simple and well-defined process, and a more traditional approach may be more effective, requiring careful consideration of options to choose the best solution.