ShraddhaLamba

The example here we want to consider is for healthcare industry and in the medical field where AI agent helps with the diagnosing medical conditions based on patient data. On one instance AI incorrectly diagnoses a patient with a different illness then actual one leading to incorrect treatment. Responsibilities to be assigned as below : To the AI Agent : Evaluate that data base and give the right set of information, if not found the relevant information then to escalate to Human intervention. To The AI Designer : Check the AI codes and algorithm at a regular frequency and ensure to the update the trained models. To The Human Reviewer: Timely Audit of the AI agent, ensuring all the compliance and security policies are abide to in order to avoid any unnecessary escalations.

Any situation of the critical areas of human being - e.g. prescribing any medicines basis the symptoms are shared by the patient. In such cases , there will always be fear and hesitance to give the complete control to AI agent. To gain the trust AI should first be enabled in a limited capacity & for simpler transactions, where the model is trained to provide suggestions and options over concluding decision. Over the period it should learn from Human decisions and make it a more robust and trustworthy to deal with

The one situation that I can think of is a credit card Process. While a customer calls the customer service for certain information regarding the credit card most common questions like validity, limit, usage, billed and unbilled amount can be answered by AI bot , even some amendment requests can be take over by AI. However, at certain point on dispute management there is Human intervention required. The signals or the criteria that AI can follow are some key words like not satisfied with the answer, unacceptable, further escalate to higher management, report on social media or legal website etc. With such key words AI should escalate this further for human intervention and validation of response before communicating the customer

For any kind of data classification & data security related tasks I would trust AI completely as in such manual tasks there are more chances of Human error always and AI has proven the expertise and best results in these areas over the period with good data set and training models. Where I would not trust AI as much is in employee performance management & calibrations with stakeholder, these areas are quite subjective and requires human intervention, emotional intelligence and interpersonal conversion. In these areas there are high chances of AI failures only sharing the logical outcome or at the best basis the data on being the models are trained on which can be limited. While in these areas of work there are endless scenarios and situations which calls for unconventional approach and skills to deal with it.

Currently for any budget approvals, we refer to the budget files manually for the availability of it & along with that few other details to ensure the expense if justified. Then we go for round of approvals on emails. What We would love to see as AI involvement here :1) Is to have the information handy on the budget availability 2) To give logical explanation on acceptance or rejection of the budget request basis pre-defined combinations 3) To beforehand share any risks/threats with that transaction