How to Lead an Artificial Intelligence (AI) Project in Your Company?
We are sometimes asked about the best way to lead an Artificial Intelligence (AI) project in a company. It is true that the AI often seems to leave the audience of its deployment on the side of the road, dumbfounded them by what they read here and there about it. AI would be on the way, like an unprecedented and unparalleled backdrop, to simply change the world.
The arrival of Artificial Intelligence in companies generates a number of fears among managers as recent studies have shown. How to make AI an ally? Why do not you fear it?
The advent of Artificial Intelligence on an industrial scale raises broad debates about its impact on employment. Many employees and managers see AI as a threat. However, recent studies seem to show that AI will be just as much a performance lever for companies. So, threat or opportunity, it all depends on how to understand the AI and to become familiar with these applications as quickly as possible.
Between myths and realities: let’s stop fearing AI!
The development of AI at work raises many questions and fantasies. The economists who have studied the question are nevertheless unanimous on one point: far from being a fad, the AI is a basic trend. This means that the AI will continue to interfere in the business. The more sophisticated it is, the more important it will be.
Today, especially useful for automating repetitive tasks, it is starting to be used to detect and prevent fraud or cyberattacks:
- in one case, it can come to compete with the human;
- in the other, it allows the human tasks that he would not have been able to accomplish.
These two only examples illustrate the two impacts of AI on employment. To look more closely, in both cases, the AI can be seen as an opportunity:
- its application in the industrial sector allows a widening of the intervention perimeter of the robots: the production costs can then be, if not more interesting, at least as much as the cost of the labour of country “at low cost”, to the point of considering the relocation of the production tool, with the logistical gains it provides;
- its application in areas such as cybersecurity, or industry 4.0, will generate needs for new skills, and thus job creation.
All the factors to make AI a large-scale structural phenomenon are there: the data, the computing power and the algorithms. Therefore, the question for companies is to start to understand it, to become familiar with it, to implement it in a concrete and pragmatic way, first on a small scale before implementing it more widely. Knowing how to use it wisely then becomes a major economic issue for companies.
AI: an offensive, moderate or defensive integration strategy?
Undeniably, the leader of a sector must be in the forefront on this subject, if he wants to stay one step ahead of the competition. Startups benefit from great agility to quickly develop innovative projects and thus nibble market share. In my opinion, three approaches to the integration of AI have distinguished themselves in recent years:
- Offensive: the company is all about Artificial Intelligence; it grants a budget and a team sufficient to integrate it into all its projects. The leader believes that they must be in the front line in their sector, because there is a premium to the leader in the integration of Artificial Intelligence. Like neo-banks and cybersecurity, very advanced on this subject;
- Moderate: they go gradually, integrating AI into some important projects but not necessarily strategic. This is worth full-scale test. The results are observed; technical developments remain in standby and, depending on the strategic orientations, can be deployed on a large scale as soon as the need arises.
- Defensive: Observe too many risks being left by competitors in this negotiation and not being able to catch up the accumulated delay.
The choice of the strategy to adopt is the responsibility of the managers of the company. If they opt for a moderate or offensive approach, the use of outside expertise is essential for its deployment. To be more efficient, it is better to trust operational managers who already have experience in conducting such projects. The business and technical vision of transition managers will quickly increase the teams’ skills.
If you miss the AI train, beware of the disappearance!
Those who have not yet integrated Artificial Intelligence into their development must quickly adopt a moderate approach, at a minimum. Otherwise, they condemn themselves to disappear more or less in the long term.
