DEV Community

Arindam Mitra

Posted on May 12

Readiness of Organisation for AI Adoption

#azure #ai #learning #community

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.

CODE REPOSITORY:-

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.

DATE	TOPICS	CONTENT
09.05.2024	Recap: Microsoft Build AI Day Switzerland 2024	https://dev.to/arindam0310018/recap-microsoft-build-ai-day-switzerland-2024-2b62
12.05.2024	Microsoft OpenAI Architecture	https://dev.to/arindam0310018/microsoft-openai-architecture-2i35
12.05.2024	Microsoft OpenAI Security	https://dev.to/arindam0310018/microsoft-openai-security-5ej6
12.05.2024	Readiness of Organisation for AI Adoption	https://dev.to/arindam0310018/readiness-of-organisation-for-ai-adoption-4eli
12.05.2024	Microsoft OpenAI: Errors I Learned	https://dev.to/arindam0310018/microsoft-openai-errors-i-learned-f4k

READINESS OF ORGANISATION FOR AI ADOPTION:-

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.

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:-

7. Use of Azure services which are AI READY for enterprise application :-
Below is just one of many example.

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: https://www.linkedin.com/in/andreaskorczak?utm_source=share&utm_campaign=share_via&utm_content=profile&utm_medium=ios_app
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:https://www.linkedin.com/in/satvinder-s-singh?utm_source=share&utm_campaign=share_via&utm_content=profile&utm_medium=ios_app
Read the Lessons learned from LSEG and Microsoft: https://www.microsoft.com/en-us/industry/blog/financial-services/2024/05/09/3-lessons-in-financial-services-ai-transformation-from-the-lseg-microsoft-partnership/
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: https://www.linkedin.com/in/ericsonchan?utm_source=share&utm_campaign=share_via&utm_content=profile&utm_medium=ios_app
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: https://www.whatisdevrel.com/

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