Three Ways To Implement Generative AI For Your Business

The first question with any technology investment for business (including Generative AI for business) is usually “what’s it going to do for me?”

The second question is always “what does it cost?”

And, as with a lot of technology investments, the answer is: “it depends”.

These days, there are three main ways to experiment with generative AI for your business. While there are a lot of resources covering features and performance of each, this article provides an overview of the commercial dimensions of these approaches. Let’s take a look.

The Easy Route – Free and Built-In

Many software companies have been rushing to integrate AI capabilities into their tools. You’ve probably seen the AI stars already popping up in your existing Microsoft or Google applications, offering capabilities ranging from autocomplete, to auto-writing, to auto-labeling and analysis. Microsoft has also been deploying a dedicated Copilot app that lets you chat with an AI right from your desktop. This trend will likely continue, and work it’s way from the biggest players down into smaller, more specialized systems in the near future.

Key benefits:

1. Price – These features are usually free (for now).
2. Integration – They are already built-in to the existing tools. Just continue to use your software normally, and get focused benefits of AI directly in the applications you already use.

Key challenges:

1. Limited Flexibility – You get exactly what the company offers, and that’s it. Options for customization, expansions, integrations with other systems, etc. are very limited.
2. Data Privacy – The data put into these systems is usually sent to the company’s servers, meaning their data security and use policies must align with your company’s needs.
3. Support – Although they come from major vendors, the fact that these tools are free means that support will may be limited to online knowledge articles.

AI As A Service

Companies like OpenAI and Anthropic offer AI as a service, charging for use of their systems on a fixed-price or per-consumption basis. They also offer multiple ways to interact – via an app, through a chat in your browser, or via API (a computer-friendly standard for direct software-to-software communication). This provides a high level of control about when and how these services are used.

Key benefits:

1. Price – The services are very affordable (on the order of cents or fractional cents per transaction), and you only get charged for what you use. This is great for experimentation prior to full deployment.
2. Flexibility – While the services themselves are limited to text or image-based inputs and outputs, there is significant flexibility to how they can be incorporated into your business. If you have unique needs or workflows, you’re not constrained by what other companies have already built.
3. Support – Since these are commercial vendors with AI as their main product, there are multiple levels of support available. OpenAI even has an Enterprise tier available offering greater performance, support, and privacy.

Key challenges:

1. Integration – Using these systems via app or chat is straightforward. However, integrating these systems with your other software or workflows requires a medium-to-high level of programming skill. You’ll need technically-skilled staff for more complex use cases.
2. Data Privacy – While you have greater control of the data shared, your data still gets sent to the company’s servers. You must ensure that their data security and use policies align with your company’s needs. These companies are also much younger than the Microsofts of the world, so their privacy and security credentials are not as proven.

AI For A Capital Budget

Along with major players in AI research, a remarkable amount of AI innovation has taken place outside of commercial companies and is freely available from online sources. For most AI applications, these systems can perform nearly on-par with commercial offerings. Since they are also running locally and your data never leaves your network, they also have major privacy advantages. You and your team have full control of their performance and use.

The challenge with these systems, though, is implementation. Some specialized hardware is needed (a small implementation is three figures, where a larger scale implementation is low-to-mid four figures), and the complexity of the installation and integration can vary significantly. If you have a technical resource available, this isn’t a problem. However, this could be a challenge for a less technical team.

Key benefits:

• Flexibility – The system can be designed to meet exactly your needs.
• Privacy – Your data never leaves your systems, providing a major advantage to data security compliance.

Key challenges:

• Implementation – You will likely need at least one employee with high technical aptitude to implement and maintain this system.
• Pricing – Between hardware and labor, you will pay more for this implementation than the others. The amortization period for this options is longest among the three.
• Licensing – There are an enormous number of systems available under varying license terms, some allowing commercial use while others don’t. You will need to ensure the system you select allows for commercial usage.

The Right Generative AI For Your Business

If speed or ease-of-use is your main focus, using Microsoft or Google’s built-in tools is the easiest way to get started.

If you want a spend-as-you-go option with high performance and support, options from OpenAI and Anthropic are terrific.

And if privacy and flexibility are your top concerns, open-source AI deployments are your best choice.

Regardless, it’s important to recognize that the AI market is developing very quickly, with significant improvements in tools, pricing, and integration coming monthly. Your focus should be on how to use Generative AI for your business, without getting caught up in the specific implementation. The best results – and biggest benefits – are yet to come.

Want to stay informed on the most important AI news in just 30 minutes a month? Sign up for the Executive Summary newsletter by AI For Business.

If you liked this post, have future posts delivered straight to your inbox by subscribing to the AI For Business newsletter: https://buttondown.email/AI_For_Business. Thank you!