The article Centaur: German AI model can predict human behavior first appeared at the online magazine Basic Thinking. You can start the day well every morning via our newsletter update.
Researchers at the Helmholtz Center have developed a AI model called Centaur that can think like a person. It is intended to predict human behavior in order to help better understand our perception.
Artificial intelligence is almost omnipresent. The technology is already used in many areas of life and provides important knowledge for research. Scientists from the Helmholtz Center for Human Center Recently developed a AI model with Centaur that is supposed to predict human behavior.
The researcher used natural language input for her system to define possible actions. The Centaur training data set is called Psych-101 and comprises ten million different possible decisions from 160 psychological experiments.
Loud official information This can predict human behavior even if tasks or context are unknown. Centaur should also be able to predict how long someone needs to make a decision.
Centaur: AI says human behavior beforehand
The researchers speak of giving them their model on humans. This would resemble the processes of the model to those who take place every second in the human brain. At the same time, Centaur simulates various decision -making processes and the resulting results. The results are not corrected, but serve as a new starting point.
As the basis for the AI, the scientists brought results from psychological experiments. The calculations are based on typical rules, which we usually use intuitively to get the best for us. Centaur also reproduces the diversity of human decision -making strategies instead of just committed to one method.
As a result, the model can make decisions based on planned thinking (model -based), as well as those that are based on learned experience (model -free) – and even a combination of both.
Meta Ki-Lama as a basis
Behind Centaur is the model Llama 3.1 70b developed by Meta. It exceeds established cognitive models in the prediction of human behavior in all 16 examined areas. The model also shows a significantly better match with FMRI data than comparable models. This applies to decision -making tasks (such as two -stage tasks) and language processing.
Nevertheless, the generalizability of the model is currently severely restricted – especially with regard to culturally different decision -making patterns. Because the underlying data have a strong focus on the western world. This is explicitly recognized by the researchers as a weakness. Nevertheless, Centaur moves closer to humans than other voice models.
Also interesting:
- Find archived chats at Chatgpt – that’s how it works
- Study: How artificial intelligence can become more sustainable
- RTT: Send real -time messages while making calls – that’s how it works
- Chatgpt: This is how you can create your own GPTs
The article Centaur: German AI model can predict human behavior first appeared on basic thinking. Follow us too Google News and Flipboard Or subscribe to our update newsletter.
As a tech industry expert, I find the development of Centaur, a German AI model that can predict human behavior, to be both fascinating and potentially concerning. On one hand, the ability to accurately predict human behavior could have a wide range of applications, from improving personalized marketing strategies to enhancing mental health interventions. However, this technology also raises important ethical questions about privacy, consent, and the potential for abuse.
It is crucial that developers and users of Centaur prioritize ethical considerations and ensure that the technology is used responsibly and in accordance with legal and ethical standards. Transparency, accountability, and robust data protection measures must be built into the design and implementation of Centaur to safeguard against potential misuse and protect individuals’ rights and freedoms.
As this technology continues to advance and become more widespread, it will be important for regulators, policymakers, and industry stakeholders to collaborate on establishing clear guidelines and regulations to govern its use. By working together to address the ethical and societal implications of predictive AI models like Centaur, we can harness the potential benefits of this technology while minimizing the risks and protecting individuals’ rights and well-being.
Credits