3 answers
Emily
Student
Overland Park, Kansas
2
Questions
Asked
336 views
What is AI and automation risk in Consulting How can I get involved??
I am curious about the different sectors of consulting. I was recently made aware of AI and Automation Risk in Consulting and was wondering what type of background knowledge can I try to gain to go into this field? Especially since AI in being newly integrated into every field I think it would be a good profession to go into. What exactly is the day-to-day tasks in this role? How would I be a better candidate?
Login to comment
3 answers
Matthew Wu
Technical Product Manager
3
Answers
Miami, Florida
Updated
Matthew’s Answer
For this role, it's beneficial to have a solid foundation in technology, some experience in AI/automation, and knowledge of compliance and regulations in the industries that interest you. Your daily tasks may vary based on your specific role, but could involve connecting with clients to understand their situation, addressing their concerns, and identifying use cases; working with your team to determine how AI is relevant and applicable, as well as any potential risks, such as regulatory issues; and creating an easy-to-understand plan to advise the client based on your findings.
To become a strong candidate, think about taking courses related to compliance and regulation, and engage in projects focused on AI/automation. Gaining hands-on experience is always a valuable asset for roles like this.
To become a strong candidate, think about taking courses related to compliance and regulation, and engage in projects focused on AI/automation. Gaining hands-on experience is always a valuable asset for roles like this.
Thanks for the advice.
Emily
David Resseguie
Data Science
1
Answer
Knoxville, Tennessee
Updated
David’s Answer
You are correct that the topic of responsible AI is super important to leaders and businesses in lots of fields, given the accelerated adoption of AI. With the great potential for AI, also comes great risks. I regularly talk to clients that want to get started using AI to improve their business, but need help understanding what those risks are and how to guard against them. Businesses need to understand what the AI is doing, and why. Is it making accurate, bias-aware decisions? Is it violating someone's privacy? The need for consultants in this space will continue to grow and would be a great career choice if you love working at the intersection of technology and business.
To be successful in this role, you would want to have a foundational understanding of the main focus areas within AI: not necessarily the underlying data science behind the algorithms, but the types of problems they are intended to solve, what types of data are used and where it comes from, and how those models are evaluated and tested. You should also be familiar with current policy and regulations that governments and other regulating bodies are issuing. It would be good to know some of the general concepts of what those policies are intended to address, though it's not expected that someone just starting their career in this space would already know the specifics of any particular regulation. You'll have plenty of opportunities to learn those as you grow in your career! And since this role will likely require significant interaction with clients, it's always valuable to spend time improving your communications skills and getting better at data "storytelling" techniques.
The day-to-day activities in consulting may vary, but it will typically consist of:
1) Staying up-to-date on the latest regulatory changes for a particular industry that you may specialize in
2) Understand a client's challenges/needs and translating those into a plan of action
3) Developing reusable frameworks and toolkits that can be used to accelerate the application of Responsible AI techniques
4) Finding ways to better communicate and translate highly technical concepts into business language that is relevant to your clients
The area of Responsible AI is a very important and growing field, so I love that you are exploring this as one of your career options!
To be successful in this role, you would want to have a foundational understanding of the main focus areas within AI: not necessarily the underlying data science behind the algorithms, but the types of problems they are intended to solve, what types of data are used and where it comes from, and how those models are evaluated and tested. You should also be familiar with current policy and regulations that governments and other regulating bodies are issuing. It would be good to know some of the general concepts of what those policies are intended to address, though it's not expected that someone just starting their career in this space would already know the specifics of any particular regulation. You'll have plenty of opportunities to learn those as you grow in your career! And since this role will likely require significant interaction with clients, it's always valuable to spend time improving your communications skills and getting better at data "storytelling" techniques.
The day-to-day activities in consulting may vary, but it will typically consist of:
1) Staying up-to-date on the latest regulatory changes for a particular industry that you may specialize in
2) Understand a client's challenges/needs and translating those into a plan of action
3) Developing reusable frameworks and toolkits that can be used to accelerate the application of Responsible AI techniques
4) Finding ways to better communicate and translate highly technical concepts into business language that is relevant to your clients
The area of Responsible AI is a very important and growing field, so I love that you are exploring this as one of your career options!
Ok thank you for this insight! I appreciate you taking the time to help.
Emily
Ankita Mohapatra
Data science consultant
2
Answers
Chicago, Illinois
Updated
Ankita’s Answer
Hi! I work in Responsible AI / Ethical AI, and can give you a flavor of the types of knowledge that is useful to me in my day-to-day:
-- Understand the model development lifecycle. Understand how we select data, clean/process, select an algorithm, train a model, test it, deploy it, maintain & monitor it... and understand what types of internal & external documentation to provide along the way (e.g., google "Salesforce model cards" and see what you find)
--As Matthew suggested, have some general idea of what regulation is in effect & also in the pipeline. This could be highly-referenced standards like NIST RMF, or draft regulation like EU AIA
--Read into how companies assess and classify risk and the concept of proportional governance. This is the global trend in terms of how we address AI risk
--Learn how companies are internally organize. You could read up on the 3 lines of defense -- in AI, that would basically be data scientists vs compliance vs internal audit
--Keep up with the news! This is incredibly important for this rapidly-evolving field. What "bad AI" examples are coming out each week, what new technology innovations are appearing, how we validate new types of models and emerging techniques for model explainability
This is a small list off the top of my head but will hopefully set you up in the right direction :)
-- Understand the model development lifecycle. Understand how we select data, clean/process, select an algorithm, train a model, test it, deploy it, maintain & monitor it... and understand what types of internal & external documentation to provide along the way (e.g., google "Salesforce model cards" and see what you find)
--As Matthew suggested, have some general idea of what regulation is in effect & also in the pipeline. This could be highly-referenced standards like NIST RMF, or draft regulation like EU AIA
--Read into how companies assess and classify risk and the concept of proportional governance. This is the global trend in terms of how we address AI risk
--Learn how companies are internally organize. You could read up on the 3 lines of defense -- in AI, that would basically be data scientists vs compliance vs internal audit
--Keep up with the news! This is incredibly important for this rapidly-evolving field. What "bad AI" examples are coming out each week, what new technology innovations are appearing, how we validate new types of models and emerging techniques for model explainability
This is a small list off the top of my head but will hopefully set you up in the right direction :)
Thanks for your encouragement!
Emily