AI developments are accelerating at a dizzying pace, and Sam altman, CEO of OpenAI, tries to look into this somewhat exciting and somewhat disturbing future. During a recent event, I talked about it here, Altman revealed o1, a new model of artificial intelligence that promises to overcome the limitations of its predecessors. What makes o1 so special?
The answer lies in this new generation of AI’s ability to “reason” through complex problems, opening up possibilities that never existed before. From revolutionary medical treatments to unimaginable scientific breakthroughs, o1 and upcoming models that work like it could be the key to unlocking the next chapter in human progress. But with great power comes great responsibility…
The Breathless Race of Artificial Intelligence
- AI developments are about to make a leap forward that could redefine our relationship with technology. Sam Altman “sees” a future in which artificial intelligence does not simply answer our questions, but reasons, ponders and solves complex problems like a human would.
The GPT series models were amazing in their 'system 1' thinking, but what we wanted
they were systems capable of reasoning.There is so much value in AI being able to reason through problems: o1 is the first system that can do complex reasoning, and if you give it a challenging problem you can get amazing results.Sam altman
In practice, it is about moving from a virtual assistant that provides you with pre-packaged answers to a real “intellectual collaborator”, capable of analyzing problems from multiple angles and proposing innovative solutions.
From ChatGPT to o1: a new paradigm
The transition from models such as GPT-4 a o1 represents much more than a simple update. It is a paradigm shift in the way AI “thinks”. While previous models, often mistakenly or “hallucinating”, processed information token by token (for those who are just tuning in now, I explain here what a token is1), o1 introduces a reasoning-based approach.
Altman likens this leap to the move from GPT-2 to GPT-4, but emphasizes that we are only at the beginning of this new phase. “Think of it as if we are at the GPT-2 stage of these new models of reasoning. You will see it catch up to the GPT-4 models in the next few years,” he said.
The promises of AI developments
But what are the practical implications of these? AI developmentsAltman outlines scenarios that seem to come from a science fiction novel:
- Healthcare revolutionized: Imagine AI capable of analyzing complex data and the complete medical history of each patient, to propose personalized treatments or discover new ones;
- Tailor-made education: Each student could have a personalized virtual tutor, able to adapt to his or her learning style and specific needs, to support and integrate the activity of real teachers;
- Scientific discoveries: AI could accelerate research in fields such as clean energy, climate change, human longevity;
- Technological innovation: AI systems could help design new technologies, pushing the boundaries of innovation.
AI Developments, Ethical and Practical Challenges
However, as mentioned, with great promise comes great responsibility. AI developments raise a number of ethical and practical questions that we cannot ignore:
- Privacy and security: sorry to harp on the point, I always do, but how can we ensure that such powerful AI systems do not compromise our privacy or be used maliciously?
- Technological inequality: Is there a risk that these technological advances will widen the gap between those who have access to these resources and those who are excluded?
- Impact on work: How will the job market change with the advent of increasingly capable AI?
- Control and responsibilityWho will be responsible for the decisions made by increasingly autonomous AI systems?
The future could surpass us
Altman points out that, although we are at the dawn of this new era of AI, the progress is tremendously rapid and tangible.
You will see updates soon as we move from o1-preview to o1. The improvement curve is very steep, and problems that models cannot solve today will be solved in a few months, not years.
Sam altman
Preparing for the Impact of an AI-Driven Future
Given the speed with which the technological frontier is continually shifting, it is essential that society as a whole prepares itself to face the challenges and seize the opportunities that these AI developments will bring.
This will require mainly 4 things: a informed public debate on the ethical and social implications of advanced AI, massive investments in training and retraining to prepare the workforce for the future. Collaboration between technology sector, government and society civil society to establish appropriate guidelines and regulations. And most importantly: continuous research not only on the technical aspects of AI, but also on its social and psychological impact.
AI developments: a future to co-create
- AI developments that Sam Altman and OpenAI are pursuing represent a seismic shift in the field of artificial intelligence. By moving from systems that simply process information to systems that reason and solve problems, we are entering uncharted and potentially revolutionary territory.
However, the future is not written. And we cannot delegate this to artificial intelligence. It is our duty to guide these developments ethically and responsibly to prevent AI developments from creating new, major problems. As a society, we have the opportunity and responsibility to co-create this future, ensuring that the benefits of advanced AI are distributed equitably and that risks are managed wisely.
And this is especially true for Altman and those like him.
- A token is the basic unit with which AI models like ChatGPT process text. It represents a part of a word, a whole word, or a symbol, allowing the model to understand and generate language. ↩︎
