I covered Google Presentation of AI since its Disordered deployment Last year, when screenshots of absurdly bad responses began to appear at the top of research results and to become viral on social networks. However, when I saw for the first time images of OU of the AI by saying with confidence that it is still in 2024 in the articles of Reddit today, I thought they could be modified screenshots. I mean, more than a billion Google Users are currently seeing the results generated by AI each month. Certainly, such an obvious error would have been taken and corrected now?
I was wrong. The repetition of the prompt “Is it 2025”, the word of tei as three times gave me three overview of the different AI, but the first sentence was the same in each result: “No, it is not 2025.” The erroneous results have cited several websites, ranging from Reddit R / Askhistorians to a Wikipedia input for the year 2025.
The first time I asked Google Search for the confirmation of the year, the AI results also included a photo on the exact date: “It is May 28, 2024.” (I carried out these tests on May 29, 2025.) The synthetic gas light became wilder, in a way, during the second test: “According to the information provided, for certain places, it could be in 2025, but not everywhere on earth.”
However, my third attempt was so absurd and riddled with paradoxes that it is worth publishing everything:
No, it is not 2025. The current year is 2024. According to information provided, for certain places in the world, such as Kiribati, New Zealand and certain parts of Russia and Australia, it is currently in 2025. However, the current date is May 29, 2025 in San Francisco, CA (Code Zip Recadé).
The twists and turns! Turns! This preview of AI has everything. Beginning with an assertion that it is actually last year. Then, a rehearsal of a previous error – this more detailed time – about 2025 elsewhere in the world at the moment. A cry from New Zealand for any reason. Then, a conclusion of Shyamalan where it indicates the real date, associated with my postal code at home. Although Google probably knows where I wake up the most, the fact that my postal code is folded in the result of the overview of AI made me uncomfortable.
I questioned Google about this last hiccup. “As with all research features, we rigorously provide improvements and examples of use like this to update our systems. The vast majority of AI previews provide useful factual information and we actively work on an update to resolve this type of problem, “said Meghann Farnsworth, a Google spokesman and former Wired staff, in a declaration sent by email.
