**12. Conclusions**

As shown the 2012 and 2013 winners of the Loebner Prize can easily be disclosed as non-human as they display almost zero consciousness and sense of dialog cohesion. The many analysis dialogs with GPT showed that it can respond usually correctly where the said champions failed to. However, analysis dialogs with more abstract areas, some limitations of GPTs were shown as expected since they use statistics and lack full logical reasoning. In [16–20] their limitations for intelligent question answering and similar tasks are examined.
