Measuring your Generative AI Maturity

Understanding your organisation’s Generative AI maturity is fundamental to planning how you can successfully adopt AI.

October 9, 2024

What is generative AI maturity?

Generative AI (GenAI) maturity measures how evolved your organisation is in its procurement, deployment, and application of GenAI to augment and improve existing corporate workflows.

The value in knowing and understanding your organisation's generative AI maturity means better aligning AI with your business goals and detoxifying stakeholders from rampant AI hype. It's also a helpful measure to establish AI governance and skill up your teams.

The generative AI maturity curve

While it seems that every enterprise wants to take advantage of GenAI immediately, it takes time to put in mature frameworks, tools, and processes to ensure that GenAI is benefitting the business and receiving the desired return on your investment as it advances in your enterprise.
Here's a typical GenAI maturity curve:

Level 1: Ad hoc

The first level of GenAI maturity is characterised by a lack of formal concern or mandates about security or risk management. One of the unfortunate signs of Ad Hoc Generative AI maturity is the potential for your corporate data to find its way into the public training data of OpenAI's ChatGPT or Anthropic's Claude AI because employee-driven experiments and tests dominate this phase.

Also, AI tool choice begins during this phase. For example, users might lean towards the free version of ChatGPT by taking the shadow AI route. That can lead to ChatGPT Enterprise becoming a top AI tool candidate for corporate adoption.

An Ad Hoc phase doesn't necessarily need to be all about data security and risk. Your enterprise can get ahead of Ad Hoc GenAI adoption by proactively issuing AI policies to your employees even before the official enterprise adoption of AI tools. Corporate-sponsored testing of the free version of ChatGPT on some of your more pressing use cases would occur during this phase, albeit with sanitised corporate data, to prevent the scraping of sensitive corporate information for an AI vendor's training data.

The exit criteria for this phase often vary depending on the organisation. Enterprises usually leave the Ad Hoc phase when GenAI becomes integral to customer-facing work or significant corporate initiatives.

Level 2: Limited adoption and innovation

The next maturity stage is limited adoption and innovation, triggered by an Ad Hoc discovery or a formal initiative. This phase marks the initial stakeholder support for GenAI adoption, involving budget, IT, and security team support. Hallmarks of this phase include pilot projects that will eventually scale out to the whole organisation or department-level applications of GenAI that may not see mass adoption.

AI tool choice advances to at least some finalists as developers and stakeholders make their tool selections and cut their first purchase orders for AI tools during this phase.

During this phase, pilot projects begin on a small scale and expand gradually, often at the team or department level. It's typical for security and compliance concerns to persist during this stage, as security and operations teams use these smaller pilots as testing grounds for policies and governance standards before the new technology is universally adopted.

Exiting this phase means a successful outcome for the one or more pilots that took place. AI reality begins to overtake hype during this phase as teams start to see the results of initial GenAI implementations for themselves and the business value they represent.

Level 3: Whole enterprise adoption and governance

The whole enterprise adoption and governance maturity phase sees teams implementing GenAI tools in production with security and governance controls in place.

Formal ownership over AI initiatives occurs during this phase, often involving IT leaders and business unit heads. Larger organisations may hire a Chief AI Officer.

Formal ownership of AI initiatives begins during this phase, often involving IT leaders and business unit heads. Larger data-intensive companies like financial services firms may hire a Chief AI Officer during this phase.

Growing up quickly

GenAI maturity continues after these major phases as your organisation learns more about new GenAI models and tools. Growing maturity also requires increased operational scrutiny as more stakeholders seek to justify the ROI of GenAI and monitor ongoing concerns about AI ethics, security, and compliance.

Narus logo