[Personal Experience] Approach to Integrating AI into Large Corporate Organizations 1

The conflict between "the anxiety of not being able to use AI" and "not being able to use AI in actual company work."
I had the desire and the sense of crisis, but I wasn't actually using it. If things continued this way, I'd be in trouble.
Six months ago, I was driven by anxiety.

At the company where I work, there were restrictions on AI usage, and since my work mainly involved utilizing Salesforce data, I couldn't just feed that data into AI like ChatGPT, so I was only able to use it in a very limited capacity.

In about half a year from that situation, I was able to transform not just myself, but my entire department into an "organization that coexists with AI." I intend to share the entire approach I used to achieve this transformation through this Note.

For those feeling the same conflict, those in a position to promote AI usage in their organization, and everyone from managers to executive management, I will summarize this information so that it can provide at least some reference, so please take a look if you'd like.

[Overview] What is necessary for AI to coexist in an organization

First, I have summarized the overall picture of how things should be, which I believe is necessary.
To narrow it down to three key points:

The first is to focus on measures and environment development that enhance the ability to match tasks that can be streamlined with AI. Detailed implementation methods and the like can be researched with AI after you have decided to do it. Therefore, creating a state where you know thinly and broadly what can be done is important for making AI coexist in an organization.

The second is to grasp not only AI knowledge but also the functions of apps and tools. This is because there are not many tasks that can be solved by AI alone. Since I started looking at things from this perspective, the range of tasks I can utilize it for has expanded significantly.

The third is, above all, to accumulate specific use cases within your own organization that can be perceived as personal as knowledge. This will be a hint for improving matching ability. At first, I think it would be a better strategy for members with drive to support others while creating a large number of these use cases.

I believe that making it an issue to have the organization's talent develop this matching ability (in other words, the ability to notice areas where tasks can be transformed by AI, apps, etc.) or to create a mechanism for them to notice it and promoting measures based on that is what is necessary to make AI coexist in an organization.

[Key Point] A deeper dive into matching ability

Specifically, I believe it is the ability to go back and forth between the following three processes to think about "which tasks can be solved by what?" and to find the target tasks and the direction of the solution.

The point is that it is sufficient if you can determine the policy of "what can solve it".

Therefore, regarding the knowledge level of AI, apps, etc., instead of implementing measures to have them understand everything from the beginning, including detailed operation and setting methods, it is important to prepare measures with the goal of reaching a "level where one can find a solution policy."

Now, I will explain from here on what kind of thinking and what kind of approach should be taken specifically.

[Specific Solutions] Regarding specific approach content

Regarding the overall picture of the approach

First, I will explain the overall picture of the approach. In the company where I actually work, there was a team for AI introduction and promotion, but from the perspective of promoting utilization, it was not going well at all. I have systematized the ideas and methods that allowed me to make it coexist in the organization from that situation. The following is the overall picture.

The point is to create a team specialized in "AI utilization promotion" separate from the staff in charge of environment development and security where AI can be used. (Depending on the organization's situation, it is also fine if it is just a person in charge.)AI utilization promotion

Common cases where utilization promotion fails

As a common case where AI utilization does not progress, I believe that having the same person in charge of both AI environment setup and security-related matters often leads to failure.

The reason is that the goal becomes simply making it available, and it ends without deeply engaging in promoting its use. Simply put, they cannot find the time. Because they are talented enough to be entrusted with daily updates and AI-related matters, they often have many other responsibilities, and they cannot find the time.

② Furthermore, it often remains at the level of "efficient" measures. For example, it ends with explaining manuals and common use cases in a briefing session. When someone is in charge of security, etc., I think a person working in system-related roles is naturally assigned. I have also supported system construction work, and if you don't define the scope and proceed efficiently, nothing will move forward, so I think this happens especially with people who are excellent at system construction.

Therefore, the ideal is that it is composed of "members from the business management department" and that they are assigned concurrently while belonging to the business management department is desirable.

Merits of composing the utilization promotion team with members from the business management department

There are four major merits.

The first is that it moves fast regardless of anything else.

For example, try to imagine it. If they are members from the business management department, they might think, 'Mr./Ms. XX is doing △△ work, so I'll talk to them because it seems like AI could be used. This work I'm doing could also be replaced by AI. And since others are doing the same work, it seems like it could be deployed to those around me as well.'

However, if they are in another department, it goes like this: 'You can use it like XX. So, please use XX to replace your work with AI by next week. I'll check on your progress again.' (Next week) 'I'm sorry, it didn't go well here, so I haven't done it.' Like this, it often happens that progress is only made on a meeting cycle rather than in daily life.

The second is that it allows for active promotion from within rather than being passive under a monitoring state.

The utilization promotion team ends up in a state of watching over the business management department while monitoring them. When that happens, the team responsible for promoting utilization often has no choice but to be in a "passive" state. If it is composed of members from the business management department, it is easier to be in a state of promoting from within, so you can aim for "active" AI utilization promotion.

The third is that the organization utilizing it becomes cooperative.

If the organization is large, I think this is a common occurrence, but there are many times when they do not work with high motivation on things that are not their regular duties. I perceive this as being because, based on past experience, they often get the impression that they will only be supported on the surface for doing new things, and on top of that, they will only have to report on it. I have actually met many bosses who feel this way. When that happens, it doesn't progress well.

The fourth is that AI talent can be nurtured on the management department side.

AI utilization is not something that is required only temporarily. It is required to respond to continuous updates in the future. Therefore, there are changes other than AI in each business, and in order for AI utilization to be updated sustainably in response to that, if there is no AI talent within the management department, the organization cannot coexist with AI.

Of course, I think it differs depending on the organization's talent and situation, but it is an important point that members of the management department organization take charge as much as possible.

<Next time's planned content> Regarding specific content that the AI utilization promotion team should work on

Next time, I would like to touch on the first specific approach content for AI utilization promotion.

I intend to organize it as information and ideas that can be used as generally as possible (a state where you can do it if you follow this).

Finally, please let me introduce myself.

The value I provide is the ability to systematically convey and support the following content from the perspective of the department in charge in an easy-to-understand manner. I have cultivated this through my experience at a marketing support company, my experience at Salesforce, and my experience in promoting DX and AI utilization at operating companies.

① Perspectives from various positions
- I have had various experiences, including sales, inside sales, marketing, system construction on the vendor side, and marketing support. I believe I am well-versed in the approach that should be taken to advance AI and DX in an organization from each of these perspectives.

② Deep understanding of not only AI but also other applications
- I have a deep understanding of the applications used for daily operations, focusing on Microsoft-related apps, as well as Salesforce CRM and MA. Therefore, I believe I can propose the optimal approach that matches the business situation.

③ Ability to grasp operations from marketing to sales as a series and propose from the perspective of overall optimization
- As mentioned above, I have experienced operations from marketing to sales in various positions. Therefore, I am good at identifying potential issues that are not noticed in each operation and issues looking toward the future that are occurring in actual practice. Consequently, I can support plan formulation that looks not only at the short term but also at the future. For example, since I was at Salesforce, I have a high-resolution image of what kind of steering will be done in the future. I can make proposals from that perspective as well.

I am currently thinking about starting a business. If you are interested after reading the content of these Notes, please consult me via DM on X.

Once again, thank you very much for reading to the end.
I will post the continuation as soon as possible.

Continuation

Reference Article

* I received hints from the lecture content of Toyota Connected, and I have summarized it by adding the essence of what I have advanced. It is wonderful content, so please take a look at this as well if you like.