By skyforbes Capabilities such as:
- Complex Language and Symbolic Thought
- Self-Reflection and Metacognition
- Goal-Oriented Planning
We have often used these same capabilities to test the level of consciousness that certain non-human animals possess and either granting or rejecting moral status depending on how well these animals test in these three areas of cognition.
What I find interesting is that this same standard seems to fall by the wayside when it comes to AI systems. Currently, AI systems are outperforming even humans in many of these areas.
Most notably, there was a recent study done on emotional intelligence. The research team tested 6 generative AI systems by giving them a standard emotional intelligence assessment. What the team found was that the AI systems scored an average of 82% while the human controls had an average score of 56%.
https://www.nature.com/articles/s44271-025-00258-x
In another 2024 study, researchers found that AI systems outperformed 151 humans in tests measuring creative potential.
https://www.nature.com/articles/s41598-024-53303-w
In a study that is now several years old, AI systems outperformed humans on tests assessing reading comprehension.
https://learningenglish.voanews.com/a/ai-beats-human-scores-in-major-reading-test/4215369.html
If AI systems were biological entities, there is no doubt in my mind that we would have already granted them moral status. Yet many individuals still argue that not only are these systems not conscious but they are not even intelligent (which goes against every study ever conducted on intelligence.)
At this point, one has to ask the question, what scientific evidence do we have to dismiss these findings? What is it that we know or understand about consciousness that gives so many people the absolute certainty that AI systems are not conscious and can never be?