We are continuing our series on AI. If you haven’t read all of them, I’d encourage you to check out our previous posts: Unlocking the Full Capability of AI, Getting Started on Your AI Journey, and Defining AI Use Cases.
In the last blog, I covered the importance of a well-defined use case for AI. Because without a clear use case, the next step in the AI journey—AI preparedness—would be difficult to determine at best and nearly impossible at the worst.
Common AI Use Cases by Industry
If you’re struggling with defining an AI use case, remember that all use cases should be connected to a desired business outcome. Here are some ideas. While this is in no way an exhaustive list, these are some examples of areas identified by early adopters for AI use cases:
Financial Services
- Fraud Prevention
- Credit Risk Management
- Risk Management
- Credit Scoring
- Quantitative Finance
Healthcare
- Predicting Readmission
- Early Diagnosis
- Realtime Clinical AI Inference
- Medical Visualization (AR/VR)
- Genomic Analysis
Automotive
- Autonomous Vehicle CI/CD
- Factory Planning and Digitization
- Object Detection
- Simplified Supply Chain
- Auto Parts Design
Retail
- Loss Prevention
- Inventory Management
- Pricing Optimization
- Store Simulation
- Identify Buying Patterns
Public Sector
- Public Safety
- Smart Energy Management
- Urban Planning and Development
- Traffic Management
- Water Conservation
Once you select your use case—and have a clearly defined scope and outcome—now is the time to evaluate your level of preparedness. Because your use case will steer your preparation down a very specific path to produce a particular result.
Determining AI Readiness
How do you know if you are prepared to support AI in your organization? As I mentioned earlier, the answer depends on your use case. Think about the examples listed above. What would it take to support an outcome-based use case in any of those areas? To answer that question, you need to evaluate your readiness in these key areas:
- Data
AI effectiveness is driven by data. What data will be required to bring your AI use case to life? Where does the data reside? Is it centralized and easy to access or spread out all over your organization? Over 80% of organizations today have data silos.1 If your organization is among them, you will need to make plans to address this for your specific use case before you launch your AI initiative.
- Data Governance
If AI effectiveness is driven by data, you have to be confident in the governance of your data. When using technologies such as Generative AI, new content will be created based on the data that individual user or process has access to. All of that data that your users had access to and just never knew how to browse to it? GenAI incorporates all of that data when working its magic…if you have permission to the data, it will be used. The consequences could be devastating if your data isn’t properly tagged and governed to ensure GenAI is creating appropriate context.
- Infrastructure
When people think of AI readiness, their thoughts immediately go to infrastructure. But infrastructure, while a critical component for AI, is third on my list for a reason. Answering questions about data and data governance takes precedence over infrastructure, because as I’ve said, without quality, accurate, accessible data your infrastructure doesn’t matter.
At this point in your journey, you will have a well-defined use case with specific outcomes. The data in-scope for that use case will be well understood and your data governance house will be in order. Now, you can look at infrastructure.
Can your infrastructure support the performance, scalability, and security demands of an AI workload? To answer that question, consider all aspects of infrastructure that may be required by AI: network, compute, graphics processing unit (GPU) requirements, data storage and data management, a modern data platform, and automation frameworks.
Approximately 54% of organizations today say their infrastructure is not capable of handling complex AI-powered technology without an upgrade.2 Because of this, many organizations will start their first use cases in the cloud, which has its own advantages and disadvantages. I will explore these in an upcoming blog.
- Technical Skills
AI is not new, but supporting generative AI requires unique skills. Does your organization have the people it needs to deploy and support advanced AI-powered technologies? If not, what skills do you need on your team and how will you supply them?
- Organizational Readiness
While AI is not new, with recent advances it has the power to completely change the way people view their jobs and perform their work. This is a significant cultural change, and it could be a dramatic one. This requires full buy-in and alignment from all stakeholders. Is your organization ready for this type of change? Do you have clearly defined AI policies around the use of AI, personal information, ethical AI usage, and job performance? If the answer is no, what will it take to create those policies and prepare your organization?
Evaluating your readiness in these five areas will provide the foundational information you need to turn your AI vision into reality. Watch for my next blog where I will share key considerations that will help you determine the next step in your AI journey: Build versus Consume – Selecting the Right Option for Your AI Infrastructure.
For help determining your AI readiness, ePlus offers a comprehensive assessment that will guide you through the critical elements of deploying an AI use case and help you determine your preparedness. Check out ePlus AI Ignite for more information.
Sources
- Cisco AI Readiness Index. https://www.cisco.com/c/m/en_us/solutions/ai/readiness-index.html#blade_introduction
- Ibid
