While at Learning Pool, I contributed to the design of a data maturity tool for organisations. Reflecting on that experience, the potential for a similar approach in the realm of AI is real. Given the rapid advancements in AI and its proven impact on productivity, an AI Maturity Model (AIMM) could help organisations navigate and optimise their AI adoption journey.
A maturity model has to be simple, show progression and be applicable to all organisations; schools, colleges, Universities, public sector through to large corporates.
Recent studies and data confirm that AI technologies significantly boost productivity. However, many employees resort to using AI covertly because their organisations either lack clear policies or impose unnecessary barriers. This scenario underscores the need for a structured approach to evaluate and enhance AI maturity within your organisation.
Research also shows that many are having to use ‘AI on the SLY’ as their organisation is either lax on policy or puts barriers in their way. Having an AI Maturity Model allows organisations to first identify where they are then decide where they want to be, tactically or strategically. This also gives clarity to people within organisations on where they stand in using the technology. More clarity also allows organisations to deal with issues such as lack of oversight, potential security risks and benefits.
You can categorise maturity curves in many ways but, in general, the sanctioned and pro-active encouragement and use of AI has become apparent.
1. AI on the Sly
Employees use AI tools unofficially due to absence of policies or constraints. Here we have the presence of shadow, sporadic use of AI tools without organisational support. This is not ideal but has become very common.
2. Informal Without Policies
Employees use AI tools openly with no defined policy, restrictive measures or constraints. Here we have the known but widespread use of AI tools without sanctioned organisational support.
3. Informal With Policies
AI tools are acknowledged and their use is permitted but not actively promoted. Basic AI policies are in place, AI tools are used for routine tasks, with limited organisational support.
4. Tactical Pilot Projects
AI is used in pilot projects to test its potential and viability in specific areas. These pilot projects have defined objectives often with initial investment in AI tools and training, with results monitored and analysed.
5. Strategic Implementation
AI adoption integrated into the organisation’s strategic initiatives. AI projects aligned with strategic goals, dedicated AI teams, regular training programs, funded initiatives with investment in infrastructure and tool use.
6. Deeply Embedded
AI is embedded in the core business processes and is a key driver of organisational growth. There is a pervasive use of AI across functions, continuous AI innovation, significant investment in AI infrastructure and talent, with full IT infrastructure and support.
Ethics and compliance
Ethics are largely captured within the regulatory and legal environment in which you work. Vague, exaggerated and sometimes irrelevant ethical debates, myths and beliefs can get you stuck at levels 1 & 2. At these levels users are may be ignoring regulatory and legal issues, so risky. However, at this early stage in the market the risks are small. At 3 policies force the issue and should be aligned with rules in whatever territories the organisation operates in – US/UK/EU/China etc are all very different. In many ways the higher your progress the more likely you are to have access IT and legal advice and if you have a major provider, such as Microsoft, Google or OpenAI on an enterprise basis – they do all the hard work for you.
Change management
On advice and procurement, one must be careful with arriviste consultants who are not practitioners. There are levels of expertise within AI. As for vendor products, go through your procurement process. It is not easy, as the technology is changing but do NOT lock yourself into licence agreements beyond a year and keep options open. This is not easy at enterprise level procurement.
As for internal advice and support, it is no different from any other change management issue. HR, L&D or Legal rarely drive such initiatives but they need to be consulted as part of the change management process. AI on the SLY is a way for employees to slip round HR and L&D, who may be behind the curve on productivity.
The need for formal training tends to come in at level 4. There is a danger with premature training, such as prompt engineering, where the underlying technology changes so quickly and negates what is taught.
Conclusion
Developing an AI Maturity Model can provide organisations with a framework to assess their current AI capabilities, identify areas for improvement, and strategically plan their AI journey. By understanding where they stand on the AI maturity curve, organisations can make informed decisions to foster a culture of innovation and leverage AI for sustained growth.
© Copyright PlanBLearning
No comments:
Post a Comment