Mankind May Achieve Singularity Within Only Seven Years, Trend Shows

A new metric called Time to Edit (TTE) has been developed by a translation firm to measure the time it takes for expert human editors to correct AI-generated translations in comparison to human translations. This metric could help us determine how close we are to reaching technological singularity, potentially even before the end of this decade.

If an AI can achieve the same level of accuracy in translating language as a human, it could have significant societal impacts.

The concept of “singularity” is significant in the realm of artificial intelligence.

It refers to the point at which AI surpasses human control and rapidly transforms society. However, it is challenging to predict where it begins and what lies beyond the “event horizon.” Some AI scientists are looking for signs of singularity by measuring AI progress toward human-like abilities.

Translated, a translation firm based in Rome, has created a metric called “Time to Edit” (TTE), which tracks the time it takes for human editors to fix AI-generated translations compared to human translations. The goal is to determine an AI’s capability to translate discourse with the accuracy of a human being, which is considered a significant challenge in the field of AI. The data collected by Translated over an 8-year period shows that machines are gradually closing the gap toward human-level translation quality, which could theoretically indicate progress toward Artificial General Intelligence (AGI).

Translated’s AI progress

According to Translated, it typically takes a human translator one second to edit each word of another human translator. In 2015, it took expert editors around 3.5 seconds to verify machine-translated text, but now that time has decreased to just 2 seconds per word. If this trend continues, Translated’s AI could match human translation quality by the end of the decade, or perhaps even earlier.

Translated CEO Marco Trombetti acknowledged that progress may be imperceptible on a daily basis, but over a ten-year period, the advancements are significant. He believes that this is the first time someone in the field of artificial intelligence has predicted the rate of singularity.

While Translated’s metric is a novel way to quantify progress towards singularity, the concept itself faces the same challenges as identifying AGI more broadly. Achieving human-like discourse is a significant challenge in AI research, but it does not necessarily equate to intelligence, and there is still no consensus on what “intelligence” means.

Regardless of whether these hyper-precise translators are indicative of impending technological ruin, Translated’s AI achievement is noteworthy. An AI capable of translating as well as a human could have a transformative impact on society, even if the actual technological singularity remains elusive.

Read the original article on Popular Mechanics.

Read more: Ominous Green Lasers Shot Over Hawaii Did Not Come From NASA Satellite After All.