One of the most striking trends of 2024 was the rapid advance of multimodal Artificial Intelligence. The evolution of technology has gone far beyond text processing: the latest versions of ChatGPT, such as the 4o model and the version with “advanced voice mode”, can now not only interpret and generate textual information, but also perform tasks with images and sounds. This has been a major step forward in human-machine interaction. This development has not only affected convenience services (such as virtual assistants or voice-driven applications) but has also fundamentally improved the quality of enterprise customer services and voice recording solutions. Such technologies have not only made digital interactions faster, but also more personalized, further enhancing the value of AI applications in practice. This may also contribute to making AI more attractive to more cautious market players, in addition to the early adopters.

Similar multimodal capabilities have been added to Google Gemini 2.0, which can process not only text but also video and images while analyzing data in real time. This development has a particular impact on areas such as creative industries, medical diagnostics and big data-driven market research, where such comprehensive analytical capabilities create new opportunities.

In addition to the rise of multimodal AI, Google also unveiled the Willow Quantum Chip, which debuted in Q4. This was a record-breaking achievement, as the 105-qubit quantum processor can solve complex tasks at unprecedented speed. It also predicts that the integration of Artificial Intelligence and quantum computing could bring radical breakthroughs in the future.

Another important innovation from OpenAI was the text-to-video generator Sora. This technology allows you to create spectacular animations, videos or even a creative mash-up of two different recordings based on a simple text prompt. The demos presented during Q1 have already demonstrated Sora’s ability to generate high-resolution and complex videos, paving the way for new tools in marketing, film and online education.

In 2024, another important trend emerged: agentic AI. The essence of this is that, unlike traditional reactive models, such systems can set goals and act proactively. This advanced autonomy can be useful e.g. in finance, when such models can independently manage investments based on real-time market information. Similarly exciting prospects open in the field of industry and services, where agentic AI can make processes even more efficient and, at the same time, safer – if properly configured. Anthropic also showed developments in this direction with the Claude 3 family (Claude 3 Haiku, Sonnet, Opus) announced in Q1, as the agentic approach can be seen in the way the model itself sets sub-goals for arranging and analyzing information.

In the race to become an “AI giant”, Apple also broke new ground last year with the launch of Apple Intelligence. This integrates generative AI into the company’s devices – iPhone, iPad and Mac – to personalize user experience and make tasks more convenient. One of the strengths of the solution is that the technology runs (partially) on the local device and aims to minimize data sharing, protecting the user’s personal information. For users who require more immediate and private experience, this could be a real breakthrough. And, of course, AI can now also make everyday tasks easier for Apple users, with smarter Siri commands, smarter photo management and new personalized health and fitness analytics. It’s also worth seeing how Apple Intelligence, combined with new technologies released in Q3 or Q4 2024, such as DeepSeek’s AI JanusFlow system, could offer even more creative and analytical possibilities soon.

The adoption of the EU AI Act was also a significant milestone in terms of regulation and ethical issues. The aim was to create a framework in which security, transparency and the protection of citizens’ rights are not lost in the drive for innovation. The regulation came into force on 12 July and highlights the fact that the level of development in this area now requires guidelines both at the international and national level. Symbolically, in Q4, John Hopfield and Geoffrey Hinton were awarded the Nobel Prize for their pioneering work in machine learning, formally recognizing its huge importance in the world of modern science.

Microsoft’s Copilot Vision and Google’s GenCast with DeepMind are also among the new features for 2024. The former mainly brings advances in image recognition, moderation and real-time computer vision, while GenCast is designed to provide much more accurate and reliable predictions than conventional forecasting methods (be it financial markets, weather or resource planning). Meanwhile, Figma AI (from Q2) is a testimony to the pervasiveness of AI in design and creative workflows: designers can now not only be inspired by an AI system but also receive concrete suggestions.

The rapid and diverse developments in 2024 indicate that Artificial Intelligence is truly not just another tool, but a defining element of our everyday lives and the economy in general. The sophistication of ChatGPT o1 and 4o versions, the speed of Google Gemini 2.0, the multimodal capabilities of Meta Llama 3.2, the creativity of OpenAI’s Sora text-to-video solution, as well as the innovations of Microsoft Vision, Anthropic’s Claude 3 family, and Apple Intelligence all show that we are talking about a technology that can have a huge impact on industry and society.

The explosive development of Artificial Intelligence in 2024 has not only brought new technologies and innovations but has also fundamentally changed the way we approach these changes. The benefits of AI, such as increased efficiency, automation of industrial processes and making everyday life more convenient, have become increasingly tangible. At the same time, the change in scale has also imposed significant challenges on industry and society. The creation of responsible regulation, the consideration of security and ethical aspects, and the issues of the restructuring of the labor market caused by AI are all topics with which development must proceed in parallel. The adoption of the EU AI Act, for example, shows that innovation and social responsibility can only ensure the sustainable development of AI together.

While the events of 2024 have already made it clear that AI will create an ever-widening range of new jobs, solutions, and business models, we must also recognize that the incredible pace of the field makes it difficult to keep track of everything. This article is also just a selection of the most important trends and opportunities, as the field of AI is evolving so dynamically that it is almost impossible to cover all aspects in a single summary.

While the future still holds many questions, it is already clear that AI is not just a tool, but a transformative force that will change the way we do business, industry and society. The years ahead for the development of Artificial Intelligence promise to be exciting and perhaps challenging, but one thing seems certain: it will continue to be a key driver of change and innovation.

István ÜVEGES, PhD is a Computational Linguist researcher and developer at MONTANA Knowledge Management Ltd. and a researcher at the HUN-REN Centre for Social Sciences. His main interests include the social impacts of Artificial Intelligence (Machine Learning), the nature of Legal Language (legalese), the Plain Language Movement, and sentiment- and emotion analysis.