Developing an Organizational AI Strategy

We've all heard about AI being a transformational technology. It has radical potential to change every aspect of a company's operations and its customer relationships. How should an organization position itself to be ahead of the curve and maximize returns from AI?

Firstly, an organizational AI strategy should be championed by the CEO of the organization and align with the business goals of the company.

Having the right level of support is vital. When steam power heralded the Industrial Revolution, did businesses achieve the potential of that by having a few random people set up steam engines in their factories, or did the CEO make the decision to go all in?

Choosing the right goals to unleash AI on is vital. The role of the CEO is to develop the organization’s strategy and communicate this strategy to the company and the market.

Does the company have a vision to dominate its industry? If so, the organizational strategy should be focused on how AI can help enable growth, not on incremental cost savings.

Recent research by Deloitte indicates that lower-achieving organizations tend to focus on efficiency or cost targets, while higher-achieving organizations emphasize growth goals such as improving relationships with customers and creating new lines of business. This also has the side effect of creating more opportunities for everyone in the organization.

The selected strategy should then be reflected in each individual business unit. Each area should be considering how they can be part of the company's vision and how they can use AI to transform their business line. This means understanding what transformations (as opposed to incremental changes) that they will make.

This also means asking questions. How will AI help build better relationships? How will AI change the way that work is done and remove routine tasks, and how can it make people more effective? How can AI help improve software development and make better products?

Other questions are about priority. What things should we do and what things shouldn't we do? To determine that, an organization needs to build a seriate of projects prioritized by ROI.

Each organization that wants to go on the AI journey needs to have business unit leaders who are prepared to learn about the technology. They need to be in a position to make informed decisions. They need to either hire new people or upskill existing ones and start with some small projects to develop practical skills and understanding before tackling larger transformational projects.

This means leveraging IT and data people, as well as selected external vendors. This is not the time to abdicate the responsibility to an external vendor. The organization has to continue to drive the initiative, is responsible for shaping it, and is accountable for delivery.

At present, generative AI is immature, but there are many opportunities to embrace transformational but mature commercially available AI products. One example is Scylla, which I have been involved with for several years now. Scylla is a very innovative company that focuses on video analytics for security and safety. Scylla’s software is commercially available and mature. Due to its numerous integrations with the most common VMS like Milestone, Nx Witness, Genetec, Exacq and Immix, among others, Scylla can cater to any case. Apart from flexibility, Scylla's solutions are impressively ROI-effective. Companies that adopt it do achieve transformational improvements and immediate ROI benefits. Check out our ROI calculator to see how much you can save with Scylla’s software.

Part of the AI journey is starting to select toolsets. It is best to be fluid, because your normal vendors may not have suitable solutions available. Some vendors have announced solutions, but they are either unavailable or not of commercial quality. This is an opportunity to explore the best tools and integrate them into your product. You also need to be prepared to move from one product to another as new tools are released. Fortunately, this is relatively easy.