We are at the heart of a tumultuous revolution that is being shaped daily by usage changes more impressive than the last. This revolution is not just a series of technological innovations but represents a fundamental shift in how society operates, a change that permeates every corner of our daily lives. These technological innovations have a profound, multidimensional influence on our lifestyles, restructuring social and economic spheres while reshaping cultural discourses and personal habits.
A Paradigm Shift
Among these innovations, artificial intelligence stands out for its exceptional disruptive potential. Synonymous with efficiency, speed, and precision, AI is changing the game, redefining the contours of what is technically possible, and relentlessly pushing the boundaries of our technological imagination. Whether it's diagnosing diseases, optimizing business logistics, or enhancing online interactions, AI intertwines in the fabric of our daily life and transforms our relationship with the world.
In this boiling landscape, generative AI text, such as Open AI's Generative Pre-trained Transformers (GPT), emerge as key players in this new technological era. They offer impressive perspectives in terms of productivity and efficiency. But, like any powerful technology, they are double-edged. Their use requires a deep, cautious understanding, a certain amount of finesse to fully exploit their potential while avoiding potential pitfalls.
It is therefore crucial to move beyond the simplistic idea that these technologies are mere productivity tools to embrace a more holistic and balanced vision of their use. This article proposes to shed light on this path, highlighting the different dimensions of interaction with generative AI text. Beyond a simple productivity race, it brings together reflective, inclusive, creative, and collaborative elements to sculpt a nuanced and humanistic approach to the use of AI.
The Perfect 'Prompt' Technique
Before diving further into the heart of the matter, it is essential to explain the operation of generative AIs. Fundamentally, they do not really understand interactions with users but rely on a vast knowledge base to predict what the next word might be in a given conversation. It's a bit like the autocomplete tool on Word. The quality of the generated response depends on the language model, the prediction algorithm, and especially the starting text, called 'prompt', provided by the user.
The quality of an AI's responses is strongly linked to the quality of the prompt provided. If you wish to obtain a relevant and precise prediction from the AI, your prompt should incorporate a series of instructions, corresponding to four types of information: context, need, request, constraints.
1 - Context
Start by providing information about the context or framework of your request. For example, if you are a graphic designer working on a new logo for a technology startup, your context could be: "I am a graphic designer working for a growing startup in the educational technology sector".
2 - Need
Indicate the specific problem you are looking to solve. If, in our example, you need innovative logo ideas, your need would be: "I need to design a fresh, modern, and distinctive logo to represent the company's brand".
3 - Request
This is the action you want the AI to perform. In our situation, you might want the AI to generate a few logo drafts: "Propose innovative and captivating logo drafts".
4 - Constraints
These are the limitations and specific guidelines that will clarify your expectations. Continuing with our example, your constraints could be: "The designs should incorporate the theme of education and technology while remaining simple and easily recognizable".
To improve the quality of responses, it is recommended to converse with the AI to better specify your expectations and not hesitate to give it feedback to refine the results. This conversation is actually a set of iterations, which are essential for both you and the AI. The exchanges help you better specify what you expect from the AI, and the AI each time has a new chance to propose content integrating new context elements. To get a suitable response, it is not uncommon to have to iterate several times. In this sense, it is best to avoid considering the AI's first response as the one you will retain! The more complex your request is, the greater number of iterations you should expect.
Getting Out of the Productivist Rut
AI, due to its automation potential, often finds its use in improving productivity. However, it offers much more than time savings and can present real challenges in its application. Let's transcend the purely productive notion to explore some remarkable uses:
AI, by automating repetitive tasks, is ideal for increasing productivity. For example, a community manager could use GPT to generate ideas for social media posts, a time-consuming task that can be automated. A virtual assistant equipped with GPT can help in drafting emails, creating blog content, and even scheduling appointments, thereby freeing up time to focus on strategic activities.
AI can stimulate intense reflection about ourselves and our social structures. Take the example of a sociologist using GPT to produce questions from different perspectives for a fieldwork survey. These questions, originating from an algorithm rather than a human mind, can bring unexpected approaches and allow the investigator to break away from their preconceived ideas to reveal complex societal realities.
AI, provided it is properly trained, can help eliminate certain unconscious biases, particularly in the recruitment process. A recruiter could use GPT to check the impartiality of job descriptions. By analyzing the language of job descriptions, AI can identify any gender, age, or ethnicity bias and suggest a more neutral and inclusive formulation.
Generative text AI can stimulate creativity by proposing new and unexpected ideas. For example, a product manager could use GPT to generate ideas for new product features. By entering a brief description of the product and its market, GPT could suggest features that complement or improve the product, thus stimulating innovation.
By using generative text AI, a project manager can structure his team's workflow more efficiently. For example, GPT can help in drafting clear directives, meeting reports, or action plans. Moreover, in preparing progress reports, AI can facilitate communication within the team and between different teams, thus stimulating collaboration.
It's important to remember that all these uses of generative text AI involve human-AI collaboration. AI is a powerful tool that can help us accomplish tasks, spur our creativity, and even improve our decision-making. Still, it is not a panacea or a substitute for human intelligence, expertise, or intuition. It is a collaborator, capable of assisting and complementing us. It is therefore crucial to understand its strengths and limitations to use it in the most appropriate manner.
Although AI is often synonymous with increased productivity, it also conceals a much greater potential. Capable of stimulating thought, creativity, and collaboration, it nonetheless requires responsible use to deploy its full potential. It is through our understanding of these technologies and our commitment to ethical use that we will succeed in exploiting all the promises of AI for today and tomorrow. Let's take the time to familiarize ourselves with these tools, explore and experiment, to navigate the digital era with the best assets.
[Article written on July 22, 2023, by Jeremy Lamri with the support of the Open AI GPT-4 algorithm for approximately 20%. Images created with Adobe Firefly, all rights reserved, 2023].
If you are interested in the combination of web 3 and HR, I invite you to subscribe to the dedicated newsletter that I keep writing on the subject, and to read the articles that I have written on the topic: