ARTICLE UPDATED IN MAY 2026
2024–2026: From the generative “boom” to invisible AI
In 2024, the world was taken aback by generative artificial intelligence models: tools such as ChatGPT, DALL·E, Gemini, and Midjourney captured unprecedented attention and brought AI into our everyday conversations. Millions of people began using them to write texts, create images, or summarise information in a matter of seconds.
As the years have passed, the initial euphoria has given way to a more realistic view. The “hallucinations” of the models – incorrect or invented answers that appear truthful – have highlighted their limitations and reminded us that we are not dealing with “general artificial intelligence”, but with very powerful yet specialised systems. In parallel, regulation has begun to take shape in Europe with the entry into force of the AI Act, which aims to ensure a safer and more responsible use of these technologies.
Nowadays, AI is already integrated into the tools we use daily, from email to mobile phones, and the phase of big headlines is giving way to a quieter stage, where AI has to demonstrate real impact on productivity, efficiency, and decision-making.
2025–2026: The (real) emergence of AI agents
Throughout 2025 and 2026, a new concept that goes beyond the “intelligent chatbot” has been consolidated: AI agents. Unlike generative models that only respond, agents can plan, connect with other tools, make limited decisions, and execute tasks almost autonomously. In the business environment, this translates into systems capable of chaining actions: searching for information, updating a document, sending an email, and logging a task in a project management system, all within the same workflow.
The major office suites have already adopted this paradigm. “Copilot”-type solutions are integrated into suites like Microsoft 365 or Google Workspace to draft reports, summarise meetings, generate presentations, and automate repetitive tasks directly from the applications that the company already uses. According to recent forecasts, a significant portion of corporate applications will incorporate task-specific AI agent functions before the end of 2026, progressively replacing traditional chatbots.
This phenomenon fits into the trend of hyperautomation that we are already experiencing: the combination of AI, process automation and data tools to eliminate friction in day-to-day work. Moving beyond the initial phase of curiosity, agents are becoming a silent infrastructure that frees up time for professionals and teams to dedicate to more creative, analytical or strategic activities.
Artificial intelligence: a simple definition
Artificial intelligence (AI) is a branch of computer science that develops systems capable of performing tasks which, until recently, required human skills: understanding natural language, recognising images, detecting patterns, or making decisions from data. Instead of explicitly programming each instruction, models are trained with large volumes of information so that they learn to generate responses or predictions tailored to different contexts.
Today, AI is not an isolated technology, but a transversal layer that is integrated into applications, devices, and cloud services. From the user's perspective, the important thing is not the algorithm, but the value it brings: saving time, reducing errors, personalising experiences, or enabling new ways of interacting with technology.
Examples of AI usage
1. Smart assistants and copilots: your new digital “companion”
Classic voice assistants such as Siri, Alexa, or Google Assistant have evolved into much more contextual systems, capable of understanding complex requests and chaining actions, like booking an appointment, drafting a message, and adding it to the calendar. Added to this are the new copilots integrated into work tools, which help write emails, prepare presentations, or summarise lengthy chat threads with just a few clicks.
In the professional sphere, these copilots are becoming a central piece of daily productivity. Connected to email, CRM or task managers, they are capable of proposing initial responses, extracting key agreements from a meeting or generating document drafts from brief notes. Far from completely replacing the professional, they function as an intelligent “second keyboard” that accelerates work and reduces the burden of routine tasks.
2. Smartphones and personal devices: AI in your pocket
Your smartphone is probably the device with the most artificial intelligence you use every day. Computational photography applies computer vision models to automatically improve lighting, sharpness, or portrait mode, merging several captures in milliseconds to obtain a more balanced image. Facial unlock systems, intelligent text correction, or quick reply recommendations in messaging apps also rely on AI algorithms trained with large amounts of usage data.
In recent years, “on-device” AI has gained prominence: models that run directly on the device rather than in the cloud. This allows for features such as voice note transcription, real-time translation, or photo classification by content without sending all images to external servers, improving privacy and reducing latency. The consequence is a smoother experience, with recommendations and assistance appearing exactly when you need them, without you having to think about “opening an AI app”.
3. Content generation: text, image, and video on demand
Generative models have democratised content creation. It is now possible to ask a tool to draft an article, propose variations of copy for a campaign, or translate and adapt a message to different tones and audiences in a matter of seconds. In parallel, image models allow for the generation of illustrations, photomontages, or graphics from natural language descriptions, accelerating design work in the early stages.
Cases of use in video and audio are also becoming increasingly frequent. From the creation of scripts and storyboards to the generation of short clips, AI is beginning to be integrated into audiovisual production workflows to automate basic editing tasks, subtitling, or the creation of localised versions. In this context, the key is not to replace creative professionals, but to free them from the more mechanical tasks so that they can focus on strategy, narrative, and final quality.
4. Advanced Recommendation Systems: Tailored Content and Purchases
Platforms such as Netflix, Spotify, TikTok, and YouTube already use AI-based recommendation systems to analyse what you watch, listen to, or dismiss, and from there suggest content tailored to your preferences. It’s no longer just about recommending “the most popular,” but about building a unique feed for each person, which updates in real-time according to their behaviour.
Something similar happens in e-commerce. Recommendation engines analyse your browsing history, your previous purchases and the products that other similar users have viewed, to show you items that are more likely to interest you. This not only increases the conversion rates for online stores, but also reduces the feeling of “information overload” by presenting massive catalogues in a way that is more organised and relevant to each customer.
5. Smart homes: automation that anticipates your needs
Automation has also found its way into the home, driven by artificial intelligence and the Internet of Things (IoT). Smart thermostats learn your routines to adjust the temperature automatically, optimising energy consumption without you having to intervene every day. Connected lighting systems, blinds, or plugs coordinate to create personalised scenes that are activated according to the time, presence, or even external weather conditions.
Added to this are robot vacuum cleaners, voice assistants, and other connected devices that use AI to map the home, recognise the voice of each household member, or detect anomalies such as water leaks or abnormal energy consumption. Together, the result is a more comfortable and efficient home environment, in which many routine decisions are resolved automatically in the background.
6. Smart mobility and navigation: much more than a GPS
Applications like Google Maps or Waze are no longer limited to displaying a static map; they use AI to predict traffic, suggest alternative routes and estimate accurate arrival times based on historical data and real-time conditions. Furthermore, they integrate alerts for incidents, roadworks or accidents based on a combination of official information and user contributions.
In vehicles, Advanced Driver-Assistance Systems (ADAS) support the driver with functions such as lane keeping, automatic emergency braking, or adaptive cruise control, all of which are based on computer vision and object detection models. Although fully autonomous driving still presents technical and regulatory challenges, AI is already contributing to improved safety and reduced cognitive load on long journeys.
7. Cybersecurity and data protection: intelligent defences
In a context where data is one of any organisation's most valuable assets, cybersecurity has become a key field for AI. Threat detection systems use machine learning algorithms to identify anomalous patterns in networks, access, or transactions, and generate early warnings of potential attacks. This allows for faster reactions to incidents that would be difficult to detect with static rules alone.
At the same time, AI itself poses new security challenges: from the use of generative models to create more convincing phishing emails to the manipulation of training data. Therefore, more and more companies are adopting “security by design” approaches that incorporate specific controls, audits, and tools to monitor the behaviour of their AI systems. The combination of AI-powered defences, good practices, and regulatory frameworks like the AI Act is what will allow us to harness the technology's potential without compromising trust.
A promising, demanding future full of opportunities.
Looking ahead to the coming years, the trend points towards AI that is more integrated, less visible, and much more focused on concrete results. Models will be more powerful, but also more costly and regulated, which will compel organisations to choose carefully where use cases deliver real value. Automation will continue to advance, freeing people from repetitive tasks and making space for jobs centred on creativity, critical thinking, and decision-making.
If there's one thing the Amara Law reminds us of, it's that we tend to overestimate the short-term effects of technology and underestimate its long-term impact. With artificial intelligence, this is likely to happen again: it has already transformed many everyday processes, but its influence on education, business, and employment is still in its early stages. The key will be learning to work with it, designing meaningful projects, and understanding its limitations and risks.
Training to harness the potential of AI
If you're considering a career in artificial intelligence or want to incorporate these technologies into your skillset, at IMMUNE we have training programmes designed to meet current market needs. Our Master in Artificial Intelligence & Data Science Prepare yourself to work with AI data, models, and tools in real business environments, helping you make the leap towards the most in-demand roles.
Beyond the technology itself, these programmes allow you to understand how to design impactful use cases, work in multidisciplinary teams, and adapt to an environment where AI will be a transversal skill in practically any sector. In a context where automation is growing, the real competitive advantage will lie with those who know how to combine human judgement and artificial intelligence tools to create useful, responsible, and sustainable solutions.
