Here is how they lie to you about artificial intelligence agents

Anyone tells you that artificial intelligence agents will change the world soon.

I find it funny to see everything that comes completely (again) and return to commercial operations. I cannot wait for the vision of artificial intelligence effects start using old BPM names and shortcuts to improve the organization.

PPT People, operation and technology
BlitPersons, Foundation, site, data, applications and technology
DMAIC – Determine, measure, analysis, improvement, control

The process is mining, BPMS, dynamic cases management, TOGAF, ITIL, MDM … The list is countless when it comes to the digital transformation in the old school that you must understand and be through the pain bar before you approach anywhere near the artificial intelligence screws in The mixture is above the chaotic structure.

None of these was outdated. They have been forgotten because some bright new trends came and sold as a drug treatment, and artificial intelligence agents are just another wave that does the same. I will be very afraid if CIO or CTO starts to wave the Slesforce RFP, enthusiastically.

If you do not understand how your work works at the level of the operation, how can you expect Amnesty International? If you leave a group of artificial intelligence factors to perform tasks independently, where is all the touch points defined so that you know when to stop and where the end of the end of the autonomy? Stewart Winter She participated in this earlier. It is really like inviting the theory of chaos to your company and expected to be positive.

Or do you wake up a morning while you are closed outside the system because the human resources agent overnight consider you not worthy to be the CEO because Jimmy Robot is customer service that can do better work?

You all accelerate to the bottom of a highway in a kitcy car you built and forgot to install the brakes.

Before you start talking to any “expert” or seller who pays the platform of the artificial intelligence agent, you will find your Org for this first.

At least, your business operations, work rules and alignment data must be defined to a reasonable standard that the artificial intelligence system can understand in any sense of independence And knowing where the borders and touch points are present to deliver them internally, interact with customers or clients, and how and when the process is completed instead of continuing to increase more endless tasks and procedures.

I am not associated with APQC (I spoke in their events several satellites), but they have a group of industry frameworks and contracts of work and knowledge in this field worthy of them. And if Accenture is able to use their work and pass it as their own, you can save yourself a large consulting bill and do the homework yourself.

If this page is from one of the AQPC classification frameworks look scary, it should be because of the possibility of this, This is the number of artificial intelligence agents who may run in dealing with the end of the process to the end.

There will be nothing but artificial intelligence that dances everything on everything organizes and carries out operations and tasks across the institution. The agents will align horizontally and head according to industrial or functional needs.

This also means that the multiple basic systems of the agent need to integrate in order to complete one or one process; There are already startups that disappear only in specific problems areas, but do not look totally to the largest organizational image.

So, what ends with you is not a sustainable cost of costs and productivity improvements, but a Haki structure of the platforms of the artificial intelligence agent mixed with the current systems based on a staple of another complement – Artificial intelligence coordination platforms – Once the promised land fails to hand over the sky in the form of savings).

So, again, anyone tells you that artificial intelligence agents will change the world soon. They simply do not understand the work involved in converting one process, not to mention hundreds.

Then consider this scenario: years at the bottom of the line in which your company is lined up for the merger or acquisition, and your institution is widespread with artificial intelligence agents. But this is them. Who wins?

After it has been exposed to merging and abstracts before in the huge stupid scales, it is not an easy or painful process. In fact, in many cases, wrong decisions were taken and the wrong work entity was chosen as a future country to move towards it.

Vanilla migrations do not work smoothly.

So what happens in the future when companies merge and have different agent platforms that manage the offer? Since this is or not, this is more complicated than choosing a set of applications on another, or a business operating model on another.

Now, you have to think about the data enjoyed by each of the companies and how to train the chosen agent platform and LLMS (or whatever the future incarnation you depend on), what are the operations, risks and compliance forms to choose as preparations, and how to allow integration for artificial intelligence agents in applications The remaining systems left behind (as you know, the acquired institution does not have a comparison similar to the material, and therefore it is important to work even though you do not want it …)

The presence of the AI ​​platform does not make things easier, as it makes things much worse and the possibility that it takes longer to complete it.

Some final ideas.

Apparently, “chance” is one of the most difficult words that translate into English. Perhaps this is in itself a happy accident, but what is this relationship with artificial intelligence?

Well, in an organizational context, the happy incident was used in successful Japanese companies through the ability to create knowledge not by processing information, but instead by “taking advantage of accurate and subjective ideas often, intuition, making individual employees and making these visions Available for testing and use by the company as a whole.

Now, everyone and their bridges want to convince you that you can replace people with this implicit and coincidence, with just a group of artificial intelligence agents who wander in carrying out tasks between applications. Nonaka also discussed, it will only work to process information; He will not act on intuition, nor will he create a subjective vision, nor will it take a leap of faith, and it has no intuition, and he will certainly not discover in a happy accident.

If there is anything, the artificial intelligence agent will be like a worse and most improved subject expert because over time you will find yourself pronounced “… 𝘣𝘶𝘵 𝘈𝘐 𝘢𝘭𝘸𝘢𝘺𝘴 𝘪𝘵 𝘪𝘵 𝘵𝘩𝘪𝘴 𝘸𝘢𝘺” When facing change, when you realize that removing people and paper from your organization was not a happy accident at all but with a rigid design.

Who will never wonder whether the algorithm will raise things wrong or whether this is the correct way to do business?!