Get ready for the AI revolution – phase 3 is just around the corner

According to OpenAI, we have just entered the second phase of artificial intelligence, in which language models such as GPT-4 are capable of generating natural language, assist with a wide range of tasks and are even capable of human-level problem-solving.

Sam Altman, CEO of OpenAI, has pointed out that the transition from stage two to stage three of artificial intelligence will be considerably faster than the transition from stage one to stage two. This is due to the accelerating development of AI and the exponential growth of the technology. Stage-three agent-based AI technology will play a significant role and is expected to have a major impact across many sectors. Perhaps we could even speak of an AI revolution. But what will the next stages bring?

In this blog post, we’ll look at what we can expect from the current stage of AI development, and what the next stages might bring.

Listen to a podcast in English, produced entirely by artificial intelligence, about the stages of AI development – we used only the draft of this blog post as source material. It will blow your mind! The podcast was created using NotebookLM. NotebookLM is Google’s AI-powered research and writing assistant.

How does OpenAI view the stages of AI development?

The artificial intelligence of the future is not merely a tool, but an autonomous entity capable of conducting research and engaging in in-depth reasoning entirely on its own – in other words, an independent AI researcher requiring no human guidance. As AI capabilities develop, we will see how the paradigms of software development change. AI will assist with increasingly complex tasks, enabling people to focus on more creative and strategic work. This change will lead to an acceleration in innovation and enable new solutions to be developed more quickly and efficiently than ever before.

The stages of AI development, as outlined by OpenAI, are set out below. According to OpenAI, we have just moved into stage two with the release of its new o1-preview model.

In earlier rumors about the model, it was referred to as “Project Strawberry”, which we discussed in a previous blog post.

• Level 1: Chatbots – artificial intelligence in conversational language
• Level 2: Reasoners – human-level problem-solving
• Level 3: Agents – systems capable of performing tasks independently
• Level 4: Innovators – artificial intelligence that can aid in inventions
• Level 5: Organizations – artificial intelligence that can carry out an organization’s work

Source: Bloomberg

Step 1: Discussion-based models as trailblazers

The first phase of AI development focused on language-based conversational models, such as chatbots. At this stage, AI was able to understand and generate natural language, but its capabilities were limited. Chatbots often operated using pre-programmed response templates, and their ability to handle complex questions was limited. However, the first phase was a significant step toward more advanced AI capabilities, as it laid the foundations for future phases in which AI will become even more autonomous and efficient.

Stage 2: We are already living in an era of human-level problem-solving in the development of artificial intelligence

The recently released o1-preview model represents a significant step forward in the reasoning capabilities of artificial intelligence, namely problem-solving abilities on a par with those of humans. The model utilizes advanced algorithms and deep learning techniques, which enable complex logical problems to be solved even more efficiently than before. Deep learning refers to the process whereby AI models learn from vast amounts of data to improve their performance, much like the human brain learns through experience. Advanced algorithms are models based on mathematics and computer programming that make this learning and decision-making possible.

O1-preview is therefore capable of understanding deeper connections between data and making more accurate predictions. Its enhanced reasoning capabilities open the door to new applications in various fields, such as medicine, technological development and scientific research, where the ability to process complex information and make well-founded decisions is essential.

According to Noam Brown, a researcher at OpenAI, future AI models could be programmed to think or ponder problems for hours, days or even weeks at a time. Although the technical cost of such ‘thinking’ is higher, he poses the question: what would we be prepared to pay if this were to lead, for example, to a new cancer drug or a more energy-efficient battery?

You can read the latest news about the o1 model on the OpenAI website.

Step 3: An incredible leap forward in the development of artificial intelligence

The next stage of artificial intelligence – the third stage – is what are known as ‘agents’, i.e. systems capable of independent action, decision-making and learning without constant human guidance. These agents can perform complex tasks, adapt to changing circumstances and collaborate with both humans and other AI systems. This technology has the potential to revolutionize the way we work, learn and communicate, as it enables more effective problem-solving and innovation.

The rapid development of Phase Three highlights the need to prepare for changes both in the use of technology and in addressing ethical and societal issues. It is important that both the private and public sectors work together to ensure that the benefits of agent-based artificial intelligence are realized responsibly and for the benefit of all.

A breakdown of the level of development of agent-based artificial intelligence:

1. A deeper understanding of natural language
   • In phase three, AI models will come to understand language even better, recognizing subtle nuances, sarcasm and cultural references. This will enable even more natural interaction with people.
2. Multimodality
   • In stage three, artificial intelligence is capable of combining multiple forms of data, such as text, images and audio. This enables, for example, the analysis of images and the generation of related text, or the seamless combination of audio and text.
3. Enhanced reasoning
   • Artificial intelligence is capable of carrying out more complex logical reasoning and solving problems more effectively. This opens the door to new applications, such as highly advanced diagnostic tools in healthcare.
4. Ethical and Responsible Artificial Intelligence
   • In phase three, the focus shifts increasingly toward ethical issues, such as impartiality, privacy and security. We must begin to develop artificial intelligence in such a way that it complies with ethical principles and the law.

Stage 4: The AI revolution, or innovators

With Phase Four, artificial intelligence will become even more intelligent and autonomous. This stage takes AI agents to a whole new level, where they are capable not only of adapting and acting independently, but also of generating new knowledge, innovating and anticipating complex events. AI will be able to understand its environment more deeply and make decisions based on long-term strategic objectives. This has the potential to radically transform many sectors, such as healthcare, the energy industry and education.

Phase four brings with it new opportunities, but also challenges. It is of the utmost importance to develop ethical practices relating to artificial intelligence and to ensure that the benefits of this new technology are distributed fairly and safely. Society must prepare for the changes that such advanced artificial intelligence will bring.

The level of development of agent-based artificial intelligence at stage four:

1. Intelligence close to that of humans
   • At stage four, artificial intelligence reaches a level where it is capable of performing tasks that require human intelligence, such as creativity, problem-solving and decision-making in complex situations.
2. Advanced learning capability
   • Artificial intelligence is capable of learning from even small amounts of data and adapting to new situations in real time. This makes it even more effective and applicable across various fields.
3. A deep understanding and awareness
   • The ability of models to understand context and act independently is expanding significantly. This also raises questions about the consciousness of artificial intelligence and its relationship with humans.
4. Social change
   • The development of artificial intelligence to stage four will have a far-reaching impact on society, the labor market and everyday life. It is important to prepare for these changes and ensure that this development benefits everyone.

Our experience of artificial intelligence at Hurja

We keep a close eye on developments in artificial intelligence and apply AI-based solutions in our clients’ projects as well as in our own work. With the help of AI, we have been able to improve our clients’ digital services, which has delivered tangible results, such as more efficient processes and a better customer experience. This enables us to remain at the forefront of development as progress in the third and fourth stages of AI opens up new opportunities for creating more complex and autonomous solutions.

The power of artificial intelligence in streamlining processes and services: Case studies from Hurja

Artificial intelligence enables significant improvements across various areas of business by streamlining processes and services. One example of this is the Finspection App, developed for one of our clients, which digitizes business processes such as text recognition from physical forms. The app’s further development utilizes AI technologies, such as Google Document AI and Amazon Textract, to streamline data entry and reduce errors, which further speeds up the processing of periodic inspections.

Another case study illustrating the integration of artificial intelligence is an application we have developed for our client, Spoken. The application utilizes OpenAI Whisper speech recognition technology, enabling more efficient speech recognition and text processing. The application analyzes speech data and converts it into raw text, which serves as the basis for high-quality end products, such as transcripts or subtitles. Whisper also recognizes when the speaker changes and counts the number of speakers, which improves the service’s accuracy and speed. This enables our client to offer faster and higher-quality services to their own customers.

The responsible development of artificial intelligence paves the way for tremendous future opportunities

Stages three and four of the development of artificial intelligence represent significant leaps forward in the history of technology, bringing with them enormous opportunities and, at the same time, unprecedented challenges. As agent-based artificial intelligence becomes increasingly autonomous and intelligent, its impact extends from everyday activities to critical industrial sectors. These developments enable the resolution of complex problems, accelerate innovation and open the door to new business models, the effects of which are felt globally.

The most important aspect of this development is an ethical and responsible approach. With the rapid growth of artificial intelligence, it is essential to ensure that the benefits brought by the technology are distributed fairly and safely. This requires close cooperation between the private and public sectors to ensure that AI is developed with a human-centered approach and that its risks are minimized. Society must prepare for the changes brought about by artificial intelligence, and we must create frameworks that support the sustainable use of the technology in all areas of life.

Although artificial intelligence is developing at a rapid pace, it is important to note that predictions about the future are based on current assessments. Unforeseen changes may occur in the development of artificial intelligence, and the impact of the technology may differ from what is anticipated. For this reason, forecasts should be treated with caution, and it is important to be prepared for the fact that the course of development may change. As innovation progresses, however, it is essential to remain flexible and adapt to potential new trends that may transform the world in unprecedented ways.

The future of artificial intelligence looks bright, but only if we approach it wisely and responsibly. Advances in artificial intelligence offer us the opportunity to make the world a better place – in terms of innovation, sustainable development and improving people’s quality of life.

If you’d like to find out more about how we can help your company in particular to make the most of artificial intelligence, please don’t hesitate to get in touch!