Alan Turing invented the primordial computer, and people felt so eager to imagine that the ticker-tape might come alive and have a chat that he immediately had to invent a test to stop them waffling about 'what it means to be conscious'. Nothing has ever passed this test, but chatbot companies do love to say they do, repeatedly, in myriad ways.

Hearing for the third time that year that 'Chatbot 3.5 has now passed the Turing test with an A+, far outpacing the others' felt tiresome quickly. But at least I had the comfort of Wikipedia, so I could send whatever miscreant shared this guff on the web, and show them the actual Turing Test, and how the Chatbots had passed no such test.

Now the worst has befallen us. The end is nigh. Wikipedia has begun peddling the propaganda in their blog on the Turing Test.

In the modern version, a human evaluator judges a text transcript of a natural-language conversation between a human and a machine. The evaluator tries to identify the machine, and the machine passes if the evaluator cannot reliably tell them apart.

Since nothing has passed the Turing Test, they've made 'the modern version' of the Turing Test. When I first read this, I felt so inspired by the brave, novel, methodology that I proved the Collatz Conjecture, disproved it, then proved it again, all before breakfast. While cowards huddled around the tired, old, Collatz Conjecture, I went with the new one, which I also created, and solved on the spot.

So what's this 'new Turing Test' which they solved?

Wikipedia links to an article titled:

ChatGPT broke the Turing test - the race is on for new ways to assess AI

So what does the article say about this forward-thinking, new test? It says that the title is:

The Easy Intelligence Tests that AI Chatbots Fail

What happened to the title? How strange!

But now we can get into the new test. According to the author:

Turing did not specify many details about the scenario, notes Mitchell, so there is no exact rubric to follow.

So Melanie Mitchell notes that Turing did not specify "many details" (source?) and has decided to discard whatever details he might have given, and created her own...then passed it.

It's impressive, but not as impressive as me. I just solved a Riemann hypothesis. Twice.