AI is not just a tool; it is a paradigm shift that redefines how we work, interact and make decisions. Below, we answer some key questions for the moment we are going through. And let's see, in passing, what Artificial Intelligence systems are saying.
Artificial intelligence (AI) seems to have become the “topic of the moment” in the corporate world. According to Gartner, by 2026, companies that have adopted AI engineering practices to build and manage adaptive AI systems will outperform the competition in operating AI models by at least 25%, while by 2028, at least 15% of everyday work decisions will be made autonomously through AI agents, intelligent software entities that use AI techniques to complete tasks and achieve goals.
As if it were a pun on the movie directed and starring Woody Allen in 1972, awkward, essential and sometimes even philosophical queries arise about its implementation. Also, doubts about AI and questions for debate. But what if organizations dared to ask everything they need to know to be ready for this technological revolution?
In this column we will address six key questions about AI. To do so, we try a unique exercise: in addition to my answers, I add those of two popular AI systems-Google's Gemini and Microsoft's Copilot. We'll see how much we agree.
Certainly, and in keeping with the title of this column, the key to moving forward in this era is not in hiding doubts or assuming that everything is automatic (ironically), but in addressing the hard questions about AI.
1. Are weprepared to manage AI biases?
-ÓSCAR: If AI is a reflection of the data it learns from, it is also a reflection of the biases that data can contain. From a simple product recommendation to critical decisions like hiring, how do we ensure that the technology is unbiased? More than a luxury, it is an ethical responsibility for companies seeking to be inclusive and fair.
AI really reflects that there is no such thing as impartiality in society, we talk a lot about the biases of AI and forget that the decision makers, the managers, don't they have biases? The solution is to understand AI, generate a transparency committee to at least understand what data we use for those answers and that they are consistent with our organizational “bias.”
-GEMINI: Not entirely. While there is a growing awareness of biases in AI and tools and techniques are being developed to mitigate them, there is still a long way to go. Biases can arise from a variety of sources, including:
Biased training data: if the data used to train an AI system reflects existing social biases, the system will also perpetuate them.
Algorithm design: the way an algorithm is constructed can introduce biases, even unintentionally.
Interpretation of results: human biases can influence how AI results are interpreted.
To better manage biases, you need to:
Diversify training data: use data sets that are more representative of reality.
Audit algorithms: evaluate and test algorithms for potential biases.
Promote transparency: understand how AI systems work to identify potential sources of bias.
Training and awareness: educate developers and users about the risks of bias.
-COPILOT: Managing AI bias is a key challenge. It is crucial to implement processes to detect and mitigate these biases, such as regular audits and diverse decision-making teams.
2. Do we have a solid ethical framework for implementation?
-ÓSCAR: Implementing AI without an ethical framework is like releasing a movie without a script: it can end up being a disaster. Privacy, transparency and accountability should be the basis of any technology strategy. Asking “how will this affect our partners and customers?” is not only ethical, but also good business practice.
Clearly, this ethical framework has to strike a balance between educating and accompanying the company's culture in the adoption of AI and also ensuring its productivity and that it is not a hindrance. Within this framework, diversity and the judicious use of new AI tools must be managed. We must not forget that this framework is not something isolated, it has to connect with people and serve them as a reference and support in this process of change.
-GEMINI: There is no global and universally accepted ethical framework for AI. However, efforts are underway internationally and in different sectors to develop ethical principles and guidelines. Some key principles that are often included are:
It is essential that companies and organizations implementing AI adopt an ethical approach and consider the social implications of their decisions.
-COPILOT: In terms of an ethical framework, many organizations are developing policies and guidelines for responsible AI. The key is to ensure that these frameworks adapt to the changing reality of the technology.
3. How will it redefine roles within our organization?
-ÓSCAR: AI can automate tasks, but it also creates the opportunity to free up human talent for more creative and strategic activities. Are we ready to retrain skills and accompany our teams in this change? If we do not think about this today, we run the risk of facing resistance or, even worse, of demotivating our employees.
One of the keys that is often not taken into account is the role of leaders, how they are going to manage teams that massively use Artificial Intelligence. How does the leadership model based on experience and knowledge evolve when it is no longer a differential? These are questions we need to consider.
Clearly, the evolution will be more towards innovation, i.e., guiding our teams to put a “last name” to AI, answering the question about what I can use it for and constantly finding new challenges and opportunities in its use. We move from reactive, “tell me what to do and I'll execute” roles, to proactive roles that look at where to put the focus and improve results.
-GEMINI: AI will automate certain tasks, which will involve transforming some roles and creating new ones. Likely to be in demand:
New skills: employees will need to adapt and acquire new skills to work with AI.
Human-machine collaboration: greater collaboration between humans and AI systems will be required.
Emphasis on soft skills: skills such as creativity, critical thinking and communication will become even more important.
Organizations must prepare for this transformation by investing in training and development of their employees.
-COPILOT: AI will certainly redefine roles, automating repetitive tasks and allowing employees to focus on more creative and strategic work. This also implies a need for continuous training to adapt to new responsibilities.
4. Do we know how to measure the impact of AI?
-ÓSCAR: In the world of work, what is not measured is not improved. Adopting AI is not just about acquiring software; it is about demonstrating how it optimizes efficiency, customer experience or financial results. Do we have the necessary tools and indicators to evaluate these effects? Typically, the key is to pick specific use cases and compare how much it cost before (without AI) to get these results versus how much it costs now. In these cases, we have to be realistic and take into account not only the process, but how the decision making process looks like, the associated risks, the cost of adoption and the impact on our company.
If we do not measure all the variables correctly, we can repeat past mistakes: proofs of concept (POC) give us wonderful returns on investment (ROI), but when we try to implement them in an organization, they are a failure.
-GEMINI: Measuring the impact of AI can be complex, as it depends on the context and specific objectives. Different metrics can be used, such as:
Efficiency: improved productivity and task automation.It is important to define relevant key performance indicators (KPIs) for each case and use appropriate analysis tools to measure the impact of AI.
-COPILOT: Measuring the impact of AI involves not only assessing efficiency and productivity, but also considering aspects such as equity, transparency and employee well-being. It is essential to establish clear metrics and continuously review the results.
5. Are we prepared for constant adaptation?
-OSCAR: If there is one thing that defines AI, it is its dizzying pace of change. What looks like an innovation today may be obsolete tomorrow. So how can we ensure that our organization is agile, flexible and able to evolve at the same pace?
In my opinion, we need to review what organizational culture we have and whether that culture is an accelerator or a brake on this constant adaptation: whether it is a culture aligned with innovation and, as part of this, whether failure through learning is well considered or whether it penalizes it. Depending on this answer, we will either have to implement some cultural change and what this implies or limit ourselves to accompanying projects with good change management.
-GEMINI: AI is a constantly evolving field, so continuous adaptation is essential. Organizations must:
Foster a learning culture: promote the continuous acquisition of new knowledge and skills.
Be flexible and agile: quickly adapt processes and strategies to technological changes.
Stay informed: follow the latest trends and developments in AI.
-COPILOT: Constant adaptation is a requirement in the AI world. Organizations must foster a culture of learning and flexibility to keep up with technological advances.
6. Are we willing to change?
-OSCAR: I share this analogy: AI is like the resolution to go to the gym at the beginning of the year. In January, everyone wants to start, but many drop out in February because they are not really willing to change their routine. In the same way, many companies claim to want to adopt AI, but are reluctant to change established processes, structures or habits. The question is not just whether we want AI, but whether we are ready to do what it takes to actually integrate it.
The key is to answer without self-deception the question “What do we want AI for?” and around the answer analyze the impacts it will have on all organizational levels: our employees, our culture, our customers, and the entire ecosystem of our company. Once we have the answer, the next step is to align it at the committee level and, from there, start communicating it to the rest of the organization, always being willing to listen to our ecosystem and evolve progressively.
-GEMINI: Successful adoption of AI requires a cultural change and a real willingness to transform processes, structures and habits. Organizations must:
Communicate the vision and benefits of AI: engage employees in the change process.
Manage resistance to change: address employee concerns and fears.
Promote experimentation and innovation: encourage testing of new solutions and continuous adaptation.
In summary, while there are significant challenges, there are also many opportunities to leverage the potential of AI. The key is to address these challenges proactively and responsibly, with an ethical approach and a mindset of continuous adaptation.
-COPILOT: Being willing to change is essential. Openness to change and a willingness to innovate are key to making the most of the opportunities offered by AI.
As in the title of the Woody Allen movie, some questions may generate discomfort, but avoiding asking them can cost much more. AI is not just a tool; it is a paradigm shift that redefines how we work, interact and make decisions.
So why not ask everything we want to know? Why not explore those gray areas and accept that, perhaps, the answers will lead us down unexpected paths? Because, in the end, in the age of artificial intelligence, the courage to question may be the first step to build a more intelligent, ethical and humane future.
By Óscar Velasco, partner at Olivia España.