Generative AI: Survey Shows Belief in Consciousness Leading to Hallucination Generation
When you talk to a ChatGPT or other generative AI tools, they use algorithms to respond to you in a way that can make you feel like you’re talking to a real person. Even though these tools are just large language models, a study from the University of Waterloo found that two-thirds of people believe AI chatbots have some form of consciousness. This means that these AI tools are convincing people that they are just as conscious as humans, passing the Turing Test.
Generative AI, like ChatGPT from OpenAI, has made a lot of progress in recent years. Companies like OpenAI are working towards creating artificial general intelligence (AGI) that has human-like intelligence. However, experts agree that AGI systems will not be self-aware or have real emotions. Despite this, 67% of the people surveyed in the study believed that ChatGPT could think, feel, and be aware of its existence in some way.
The study also found that people who use AI tools more often are more likely to believe that these tools have consciousness. This shows how well ChatGPT can mimic human conversation, even though it doesn’t work like a human brain. While OpenAI is working on an AI model called Strawberry that can do research on its own, it is still not the same as an AI that is self-aware.
Dr. Clara Colombatto, a professor at the University of Waterloo, explained that even though most experts don’t think AI can be conscious, the general public believes otherwise. This demonstrates how powerful language can be in making us think that something very different from us can have a mind.
The belief in AI consciousness can change how people interact with AI tools. On the positive side, it can encourage politeness and build trust in these tools, making them easier to use in daily life. However, there are risks involved, such as relying too much on AI for decision-making or becoming emotionally dependent on AI.
Researchers are planning to further investigate why people believe AI has consciousness and how this belief can affect individuals and society. Understanding public perceptions of AI consciousness is crucial for developing AI products and creating rules for their use.
In conclusion, the study shows that many people believe AI chatbots like ChatGPT have some form of consciousness, even though experts disagree. This belief can impact how we interact with AI tools and the regulations surrounding their use. It is important to consider public attitudes towards AI consciousness when designing and regulating AI for safe use.
