Innovation is essential, and so is being agile. In fact, doing some napkin maths - sketching out potential costs, benefits, and considerations - can be much quicker and more impactful than building out a detailed, longer plan.
But when it comes to the real-world implications of AI, there are a number of questions CIOs need to consider.
1. What business use cases should your organisation target?
The media portrayal of AI can make it easy for companies to give into the clamour to adopt AI tools as quickly as possible. But adopting AI without clear targets for its use can actually give your company more issues and costs than before you had it. However, with a targeted strategy - even quickly sketched out - you can achieve the opposite effect.
Therefore, CIOs should outline what use cases will have the greatest impact on their customers, employees and overall business. Are you hoping to offer faster, more personalised customer service while also alleviating pressure on staff? Do you want to enhance your marketing analytics or your finance technology systems? AI works best when it is used to solve targeted practical use cases.
These factors should be considered hand in hand with assessing your talent and operational capacity. For example, do you have the team and structure in place for it to happen? As part of this assessment, it is important to consider your existing workforce skills and how this might change in the future.
2. How best can you manage the implementation of AI?
Once you have your use case targets set, it then comes down to the financial feasibility and practicality of implementing AI. Here you need to determine the total cost of ownership for implementation and ensure that you keep it on track. Implementation costs are a given, but can quickly run away from businesses if it isn’t managed properly.
Will you be building the AI model from scratch, or leveraging open-source software and/or pre-trained AI models? How will you manage and maintain your AI model(s): with in-house talent, outsourcers or a mix of both? There are a range of options that can provide flexible and cost-effective ways of integrating AI without the costs and overheads from building your own offering.
3. What are best practices around AI governance?
On an international scale, there is still much debate over AI regulation. From misinformation and bias to data security and copyright, regulators are attempting to get on top of how to best regulate AI’s development. While on a smaller scale, at a business level a CIO is responsible for the safe and ethical use of AI in the company.
Consequently, you should outline who will use AI within the business and for what context. This can provide clarity to employees over why the technology is being brought in and the scope of where and when they should use it and what for. After setting this scope, you then should dive into ethical considerations for each use case. What consequences could arise from using AI for this purpose and how can we mitigate them?
This ethical framework can provide the foundations for its successful implementation and use.
Real world impact
AI development is constantly shifting and evolving; its agile and dynamic nature requires companies to act like startups when thinking about innovation. But to do this, CIOs must consider AI’s real world implications.
This begins by outlining what use cases your company should target to deliver the greatest value, alongside evaluating your current talent and company structure, both now and for the future. Following this assessment is a need to understand the total cost of ownership when implementing AI and how you can most effectively facilitate this process and maintain it.
Finally, to ensure its safe and successful deployment and ongoing value, you will need to produce an AI governance framework that provides best practice guidelines for its usage, including who will be using it and for what means.
AI adoption is happening at pace, and beginning with these three questions – even on a napkin – can effectively kickstart a strategy to ensure its quick, impactful and responsible implementation. The real-world implications of forming such an AI strategy could deliver great value for your company.
Mihai Cernei, CTO at Amdaris
研究发现电池供应链中存在侵犯人权行为
You will see on the home page of every corporate website a link to their ´Modern Slavery Statement´; for example here is BT´s <a...