DEV Community

Cover image for Readiness of Organisation for AI Adoption
Arindam Mitra
Arindam Mitra

Posted on

Readiness of Organisation for AI Adoption

Greetings my fellow Technology Advocates and Specialists.

This is Chapter #4 of my Data & AI Series based on Microsoft Build AI Day, Switzerland, 2024.

In this Session, I will provide readers, my views on Readiness of Organisation for AI Adoption.


Data And AI with Azure - All in One:-

Greetings to my fellow Technology Advocates and Specialists.

The Objective of this series is to learn as much as possible on Data and AI and in the process help others.

09.05.2024 Recap: Microsoft Build AI Day Switzerland 2024
12.05.2024 Microsoft OpenAI Architecture
12.05.2024 Microsoft OpenAI Security
12.05.2024 Readiness of Organisation for AI Adoption
12.05.2024 Microsoft OpenAI: Errors I Learned

1. Build the right team to bring the right motivation and productivity:-
This is the most important step but giving the benefit of doubt, it may happen that organizations are not clear on requirement or the right information from the source is lost into the translation from ear to ear.
Check the excerpt from Microsoft learn, DP-100 - Designing and Implementing a Data Science Solution on Azure. Here you will observe, how Microsoft defines who should do what. This is just an example for ease of understanding.
Image description
2. Individual productivity with the use of AI Toolings.
3. Team productivity with the use of AI Toolings.
4. Put AI capability in the tool chain.
5. Use Frameworks and not writing code from scratch:-
Concept is simple -
If you are learning, write code as much as possible.
If you are working, use frameworks and contribute to useful frameworks.
6. Enterprise Landing Zone for AI Apps:-
Image description
7. Use of Azure services which are AI READY for enterprise application :-
Below is just one of many example.
Image description
8. Talking to organisation who are ahead of learning curve with their use cases.
On the AI Day, Microsoft presented 3 short talk on Real world AI Use cases.
(A.) Andreas Korczac, Group SVP digital, Adecco group
Connect over Linkedin:
Adecoo Group created Global data platform. They are the early adopters for Microsoft Fabric and have build Custom orchestrator with LLM and container apps.
(B.) Satvinder Singh, Group head (Data and Analytics), LSEG (London Stock Exchange Group)
Connect over Linkedin:
Read the Lessons learned from LSEG and Microsoft:
The Data quality and governance process is Powered by Microsoft Fabric and Microsoft Purview with "Responsible AI Principles".
(C.) Ericson Chan, CIO, Zurich Insurance
Connect over Linkedin:
Zurich Insurance built AI assurance framework (RESTT: Reliability, explainability, safety and security, transparency)
With the use of Copilot, Zurich Insurance is reshaping insurance company.
9. AI Security Maturity:-
This must include below.
i) Understanding of Generative AI Threat Map.
ii) Security controls within AI Systems.
iii) Threat Modelling Scenarios.
iv) Threat Mapping Template.
v) Understanding of End to End Secure AI.
Important to Note: Security for AI is an ever-evolving process
10. Organisation & Cultural Change:-
a.) Unlike Platform Team (Cloud, Devops and Containers) serving needs of Organisation horizontally, So should Data and AI.
b.) Use Artificial Intelligence, with intelligence.
11. An crisp and clear TOM (Target Operating Model) to better handle data to day business.
12. DevRel: Developer relationship:-
It is the Communication between organizations and developers to ensure a better information flow and feedback loop.
More on DevRel:
13. Participating in meetups and conferences.
Sharing is caring and presenting your Solution in front of unknown audiences with their unexpected questions always results into learning and further development.

Hope You Enjoyed the Session!!!

Stay Safe | Keep Learning | Spread Knowledge

Top comments (0)