• 1 - Introduction:
• 2 - The Turing Test?
• 3 - Brave New World?
• 4 - Progress:
• 5 - The problem of Consciousness:
• 6 - Conclusion:

1 - Introduction:

1.1 - Request 1:

1.2 - Request 2:

1.3 - Request 3:

1.4 - Request 4:

2 - The Turing Test?

2.1 - Cultural Knowledge:

2.1.1 - Request 5:

2.2 - Reasoning Ability:

2.2.1 - Request 6:

2.2.2 - Request 7:

2.3 - Imagination:

2.3.1 - Request 8:

3 - Brave New World?

3.1 - Mastery of Natural Languages and their subtleties:

3.1.1 - Request 9:

3.1.2 - Request 10:

3.2 - Self-references:

3.2.1 - Request 11:

3.2.2 - Request 12:

"Cette phrase contient 32 caractères".
 ----- ------ -------- -- ----------
 12345 678901 23456789 01 2345678901

3.3 - Logic:

3.3.1 - Request 13:

3.3.2 - Request 14:

• On 03/14/2019, Emma Harula Iwao calculated 31,415,926,535,897 digits,
• On 01/29/2020, Timothy Mullican reached 50 trillion of digits,
• On 08/2021, the University of Applied Sciences of the Grisons (HES) computed 62.8 trillion digits,
• On 06/09/2022, Emma Harula Iwao, 100 trillion of digits.

3.3.3 - Request 15:

def       is_prime(number):
          if        number<= 1:
                    return    False
          for       i in range(2, int(number ** 0.5) + 1):
                    if        number % i == 0:
                              return    False
          return    True
 
def       find_prime_even_numbers(max_number):
          prime_even_numbers = []
          for       number in range(2, max_number + 1):
                    if        number % 2 == 0 and is_prime(number):
                              prime_even_numbers.append(number)
          return    prime_even_numbers
 
prime_even_numbers = find_prime_even_numbers(100)
print(prime_even_numbers)
 

3.3.4 - Request 16:

3.4 - Elementary Knowledge:

3.4.1 - Request 17:

3.4.2 - Request 18:

3.4.3 - Request 19:

3.4.4 - Request 20:

3.4.5 - Request 21:

3.4.6 - Request 22:

3.4.7 - Request 23:

3.5 - Advanced Knowledge:

3.5.1 - Request 24:

3.6 - Hallucinations:

3.6.1 - Request 25:

4 - Progress:

• Through "occasional" human interventions. However, in this case, how can we guarantee that a modification in these complex neural networks does not spread and disrupt other correct information?
• By expanding the knowledge base and validating its content. In this scenario, real-time access to the Internet would be absolutely essential.
• By changing the model. Indeed, if Bard/Gemini "experimental" is solely based on GPT [01], certain problems cannot be solved by a "simple" natural language model.
• In use cases such as "encyclopedic consultation," where it is essential to distinguish between what is TRUE and what is FALSE, it might be possible to subject Bard/Gemini "experimental"'s outputs to an "adversarial" AI (as done in the field of picture synthesis or games) that would ensure their accuracy. However, this poses a significant challenge: How can we establish, validate, and update a reliable knowledge base for such an AI? But for us, beings with consciousness, doesn't it play the role of that "antagonistic" AI? Indeed, our subconscious constantly generates information (in a broad sense). When we sleep, our consciousness is no longer active, and our dreams often have the incoherence of certain outputs from Bard/Gemini "experimental". However, when we are awake, consciousness then plays the role, if necessary, of a coherence filter, particularly sorting between the TRUE and the FALSE.

5 - The problem of Consciousness:

6 - Conclusion: