How are you approaching generative AI adoption in your organizations? Are you focusing first on quick-win use cases (like chatbots/automation) or deeper integrations into core platforms?

Artificial Intelligence (AI)Generative AI
3.5k viewscircle icon3 Upvotescircle icon8 Comments
Sort by:
Director of IT in Services (non-Government)a month ago

We have a 2-pronged approach. We definitely are exploring AI solutions to fit use cases from the business, but at the same time we have identified core capabilities that we either need now or will need and are exploring how to develop / implement those capabilities. We have a color-coded AI roadmap that shows the plans for both efforts - use-case-based exploration / development vs. capabilities. I suppose a third prong would be the low- and no-code solutions that we are encouraging our staff to explore, so that they can create their own low-complexity solutions and prompts.

Lightbulb on2
Head, Software Engineering, Cloud and Digital Transformationa month ago

We began with standalone, high visibility use cases where results could be clearly demonstrated, which helped build early momentum. Notably, leveraging LLMs with GraphRAG yielded faster and more effective outcomes, leading to quick buy-in from the leadership team. As we move toward deeper platform integrations, we recognize the need for a more robust and well-vetted architecture.

Lightbulb on3
Director of Engineeringa month ago

We began with a set of small use cases that utilized Copilot to enhance individual efficiencies. We are now moving to include more complex and integrated AI solutions that would result in medium to large-scale transformation.

Lightbulb on2
Director of ITa month ago

AI Adoption can be viewed from two perspectives, firstly employee personal efficiency by using AI tools like copilot, and secondly more collective and company based efficiency and process improvement. For both its important that the business is aware of the risk and security concerns, so whether your going for quick-wins or larger transformational initiatives this should be your number one priority. I'm sure your key stakeholder would also want to know if there is ROI for any adoption, as this is important.

Lightbulb on3
IT Director, Technology Business Management Office (TBMO) in Manufacturinga month ago

We started with quick-win use cases to drive adoption and learning about AI at the company. Shortly after, platform teams started investigating AI solutions on virtually all software that we manage and we're investigating capabilities that we can utilize for deeper integrations.

Lightbulb on3

Content you might like

Does your organization have an AI CoE?

Yes43%

No, but we plan to35%

No, and we don't have plans to21%

View Results

Has anyone executed any valuable use cases using MS CoPilot they'd be willing to share? Welcome thoughts here or links to previous posts where this might be a topic.

Read More Comments

How likely are you to use the new Azure OpenAI service?

We already have plans to use it8%

Very likely30%

Likely24%

Somewhat likely23%

Somewhat unlikely3%

Unlikely3%

Very unlikely

Unsure4%

View Results

We are in the process of implementing a chatbot for our local government site with intentions to deploy for internal use only for testing.  We are having problems quantifying is the cost around token consumption.  I know that cost varies depending on the model used, but does anyone have any useful estimators.  Consumption calculators I have found don't seem particularly useful.  What methods for estimating cost did you use?  Did you use number of searches against the site?

I am interested in people who have tried generative AI coding projects.  In your experience, are there project attributes that make generative AI coding capabilities a "good fit"?  Are there project attributes that are likely indicators that generative AI will not result in something useful?  I would think complexity of requirements is a big factor.  If the complexity of the project or its requirements is low - I assume genAI coding can get you within the ballpark.  For instance, logic stores: if your project has critical business logic spread out all over the place and not centralized I think that generative AI would not be able to make sufficient sense of that.