We have reached the FIFTH episode of this series dedicated to how to hire an AI in the company.
Today there are many decisions to make to bring AI into the company but the important thing is to decide to start practicing. You probably won't choose the right model nor the best tools. But the alternative is to stand still and let everyone else carry on.
So let's continue the journey with our 'virtual intern' and see what steps need to be taken to bring him to the company.
The Journey so far 🚀
For those who have not read the previous episodes, here is a brief summary that will guide you through the complexities and challenges of this innovative process. (I still recommend you read them if you haven't already done so: many things in this post concern reasoning from previous ones.)
First episode: The Loneliness of the Manager in the AI Era
We have explored how AI is in an unstable phase, today we can only choose the worst version of AI, preparing ourselves for new methods, tools and approaches, and setting the right expectations. Read more here.
Second episode: Hire, Not Adopt
Generative AI must be chosen carefully as you choose the right tool for each activity, just as a Boeing is not made to transport only a sheet of paper for a few meters it makes no sense to ask the AI to do things for which it is not been designed. And in the second post I give a small overview of all the use-cases that you could activate.
Third episode: How the Intern is made
We have seen how access to Generative AI models varies from simple chats to complex systems such as APIs or workflow engines such as Zapier. Find out more.
Fourth episode: Which AI Model to Choose?
We addressed the choice of the right AI model, underlining the importance of Foundation Models and giving a small overview of operating costs. Very complicated topic but worth addressing: Read here.
AI Onboarding: Fundamental Steps
Onboarding, in the context of AI, refers to the process of integrating artificial intelligence into the corporate environment. Let's try to continue the intern metaphor to understand what steps we will need to take.
This process is crucial because, just like an intern, AI requires orientation and adaptation to function effectively. AI, despite being an advanced technology, needs to be configured, fed with appropriate data and constantly monitored to ensure that its actions and decisions are aligned with business objectives.
Furthermore, correct AI onboarding helps collaborators to familiarize themselves with this new "presence" in the company, understanding its limits, potential and the best way to interact with it. This approach guarantees a productive synergy between AI and the human team, optimizing the benefits that technology can bring to the organization.
⚠️ ATTENTION Failing to onboard our AI intern is like giving a newbie the wrong map to hidden treasure. Imagine AI, full of excitement and potential, wandering disoriented through the digital corridors of your company, encountering obstacles and impasses instead of finding innovative solutions. It will end up going around in circles, creating more confusion (and damage) than results. A well-done onboarding not only puts AI on the right track, but transforms that newbie into an expert explorer, ready to discover the hidden treasures in your company's data and processes, giving greater value to all the people who collaborate with him!
1. Training and awareness
First of all, it is crucial to educate staff on the functionality, potential and limitations of Generative AI. The speed at which AI runs is impressive and the risk of having people who don't know what we're talking about is very high.
I constantly realize that being aligned with the current panorama involves a great effort that only highly predisposed individuals are willing to do.
💁♂️ As you may know, this is part of my job: organizing introductory workshops to Generative AI, in collaboration with the VJAL institute all over the world. Workshops are designed to prepare teams for this new era. I know this is advertising but please do it... if you don't like me look for someone to help you (maybe on VJAL 🙂) so that the whole team is aligned. But if you want to book a workshop with me and your team, click here.
Once the team has the basics, let them tinker around" with AI for a couple of months. Imagine this period as a laboratory of ideas, where everyone can explore, make mistakes, and above all, learn.
In the meantime, consider who else to train and whether you have managers capable of doing so.
After this phase of creative exploration, it will be time to collect all the ideas that have emerged and focus on the development of the first serious pilot project.
You'll see, it will be like opening a mystery box: some ideas will be brilliant, others perhaps a little crazy, but all precious. This approach not only stimulates innovation, but also creates a dynamic and engaging work environment, where AI becomes a teammate, becomes the friendly intern who collaborates with everyone and not just a simple tool or a "directive from the high”.
2. Identify the pilot project
At this point you will have understood what the first project to do is. I advise you to start small, with simple things for internal use.
For example, a GPT on HR processes or on some 'difficult' internal products.
Commit to ensuring that the team does a recap at least every fortnight on the experiments done with this GPT: Pros, Cons, Errors, Prompts that work, Information that is missing.
⚠️ Who should do it? The head of AI. And who is the Head of AI in the company? At this stage necessarily the CEO. Who will be able to delegate it once he understands that his is an AI Driven company.
3. Data Preparation
Corporate data must be accessible, clean and structured to ensure the optimal functioning of AI. There is so much that you have to tell a new employee about yourself. You need to tell it like it is and give them as complete a picture as possible. And you need to give them all the information they need to navigate your organization. We have already talked about it in the first episode.
It is the first most important point but it is probably too complex (boring) to tell it in a blog. What matters is that everything that defines your company is clearly available to the future AI processes you plan to do. It will be one of the most expensive and time-consuming processes.
4. Risk Management and Compliancy
It is critical to identify and mitigate risks related to the use of AI, including data bias and ethical issues, as well as ensuring compliance with data privacy and security regulations. The legal aspects surrounding AI are an ever-changing sea of conditions.
These concepts in the first phase perhaps do not require a dedicated committee. But depending on the results you manage to obtain, you will have to ask yourself many serious questions. Start spinning them in your head.
In the meantime only choose solutions that guarantee the confidentiality and protection of your company data right from the start.
5. Continuous Monitoring and Feedback
This point should be added starting from moment zero. I talked about it a little above: evaluating AI performance and collecting feedback for continuous improvements is essential. You see the questions again about who should do it 🙂
6. FAFO (Fool Around And Find Out)
Experimenting with AI in real situations can reveal unexpected potential and solutions. Don't be afraid to make mistakes, to seek new paths, to try tools that have just been released.
FAFO is an approach that we are pursuing, half-jokingly, with other VJAL's trainers. It's simply a matter of launching yourself and seeing what happens. Let's experiment with AI, put it to the test in real situations and see how it fares. It's a bit like playing with a new technological gadget - we discover things we didn't even know were possible. It might be a little a leap in the dark, but that's how you learn the best things, right?
So, let's let AI surprise us with its unexpected answers and its out-of-the-box solutions: let's FAFO!
7. Here we go: Launching Serious and Structured AI Projects 🚀
After having done all your homework and having let a few months pass (not too many, eh!), the time will finally have come to launch serious and structured AI projects. Think of this moment as the great debut of your digital intern who is hired on a permanent basis.
Now that AI is well integrated and the team is on board, it's time to take full advantage of this revolutionary technology and consolidate the role of our now ex-intern.
It's time to kick off projects that leverage artificial intelligence to optimize processes, personalize the customer experience, and maybe even discover new business opportunities that were previously hidden.
Always remembering that the final responsibility is human!
This is the moment your company begins to reap the true benefits of AI: efficiency, innovation, and a push into the future.
New Roles for New Scenarios
When you start to introduce AI into "Serious and Structured Projects" you will start to feel the need for someone who "takes care" of all interactions with AI in the company.
Below is a small and non-exhaustive list of roles that are emerging in organizations regarding AI:
1. AI Integration Manager
This role requires a combination of technical and managerial skills. The AI Integration Manager oversees the implementation of AI systems and coordinates the various phases of onboarding, ensuring that AI solutions are aligned with company objectives.
2. Data Scientist / Data Engineer
Experts in data management, analysis and interpretation, Data Scientists and Data Engineers play a crucial role in optimizing data for AI. They are responsible for ensuring there is no bias in the data and preparing it so that AI can learn effectively and efficiently.
3. AI Ethicist or AI Compliance Officer
Given the importance of considering ethics and regulatory compliance in the use of AI, this role focuses on ensuring that AI applications comply with ethical principles and applicable regulations, such as data privacy and discrimination.
4. AI Trainer or AI Coach
These professionals are responsible for "training" of AI, i.e. the teaching and improvement of AI models through feedback and training. They play a crucial role in improving the accuracy and effectiveness of AI systems.
5. AI Project Manager
This role deals with the planning and execution of projects involving AI. The AI Project Manager must understand both technical and business aspects, ensuring that projects are completed on time, on budget and to desired objectives.
6. AI Security Specialist
This role focuses on the security of AI systems, ensuring they are protected from cyber threats and that data is managed securely to prevent breaches and losses. And it will be a very long and important chapter!
Did I forget anyone? Of course, the most important:
The Head of AI
We have seen that in the first phase the role belongs to the CEO. But this too is a task that sooner or later will have to be delegated.
The Head of AI in a company plays a crucial role, acting as the main architect and strategist for the integration of artificial intelligence. This key figure is responsible for defining and implementing the AI vision and strategy within the organization, ensuring they are aligned with overall business objectives.
The role will be well defined over time. In the meantime I asked Chat GPT to propose Responsibilities and Functions and a Job Description.
Click, take a look and tell me what you think
Responsibilities and Functions
Job Description
What's missing?
I know that there are many aspects that have not yet been explored in depth such as cost-benefit analysis, the differences between large and small companies, between technology and manufacturing companies but this (for now) is just a blog and not a book on AI 🙂
Now, a question: in your opinion, what is still missing for AI onboarding? What topics do you feel should be covered more and in more depth? Tell me! I want to hear your ideas, your perspectives. After all, in this AI journey, every voice counts and can be the key to unlocking hidden potential. So, unleash your boldest ideas and deepest reflections. Together, we can build AI onboarding that not only works, but truly takes businesses to new heights!
So What…
Remember that bringing an AI on board in the company is a process that requires patience, curiosity and a holistic approach. Managers must be prepared to:
Train people on the capabilities of Generative AI.
Ensure the accessibility and quality of data.
Prepare a short-term plan with the aim of structuring a more complex long-term one.
Be aware of the risks and inform yourself adequately.
But, beyond the techniques and strategies, there is an element that I will never tire of reiterating as fundamental: the mentality with which we approach this technology.
AI is not just a set of algorithms and data, but a dynamic entity that can grow, learn and, above all, contribute significantly to the success of our company. Like a novice transforming into a seasoned expert, AI can surprise us with creative solutions and profound insights, as long as we provide it with the necessary guidance and support.
We must therefore embrace an open and collaborative approach, encouraging experimentation, continuous feedback and flexibility. It is through this mindset that we can fully exploit the potential of AI, transforming it into a valuable ally in our corporate journey towards innovation and excellence.
In the end, AI onboarding is not just a technical process, but an exploratory adventure that requires curiosity, creativity and, above all, a strong synergy between humans and machines.
Let's remember that, in the world of AI, the limit is not technology, but our ability to imagine and innovate. So, let's continue to explore, experiment and above all dream, because it is in this unlimited space of the imagination that true revolutions are hidden. 🚀🌟💡
As always, I invite you to reflect, comment and spread ideas by sharing this post with people you think might be interested.
To stay updated on my contents:
Read Glimpse, my novel about artificial intelligence. p>
Or contact me here.(Especially if you think your organization should need a pratical AI Workshop!)
See you next time!
Massimiliano Turazzini
Comments