The AI ​​large model data indicates that the monthly income of "migrant workers" is no more than 5,000 yuan, and the unit price has dropped from 5 cents to 4 cents.

IT House News on October 9th, AI large models have been a hot topic in the field of artificial intelligence in recent years. They can achieve various amazing functions, such as generating realistic text and images, or communicating smoothly with humans. dialogue. But behind these large models, there is a group of unknown data annotators who work hard every day to add labels to raw data and provide AI technology with the massive data required for training.

Picture source Pixabay

The job of data annotators is not easy. They need to face tedious tasks, low income, long-term instability and the risk of being replaced at any time. They are the cornerstone of the development of artificial intelligence technology, but they receive little attention and respect

According to the report of "Tech Planet", data annotators use the most primitive piece-rate system to calculate wages, and most practitioners earn no more than 5,000 yuan per month. Some of them are college graduates, some are mothers, and some are career changers. They process pictures, text, voice and other data in small cubicles in third- and fourth-tier cities, providing services to Internet giants and automobile companies

IT House has noticed that the data annotation industry has also experienced ups and downs. In 2017, when expectations for AI technology were booming, data annotators could earn high incomes, with a 2D pull box earning 50 cents. However, as industry competition intensifies and technology development fails, the unit price of data annotation is getting lower and lower, now as low as 4 cents.

The data annotation industry is facing tremendous pressure. The company needs to have a certain scale and capital reserves to obtain orders and deal with problems such as long payment cycle, high employee turnover, and unstable quality and cycle. Haitian Ruisheng is the first company in the industry to be listed on the main board. Last year's profit margin was just over 10%, but it has fallen into losses in the first half of this year

Data annotators are increasingly worried that they may soon be replaced by artificial intelligence they themselves helped create. Some domestic and foreign companies are developing tools that can automatically label data, using mainstream large models on the market to label data sets. These tools claim to improve annotation efficiency, reduce costs, and achieve accuracy rates that are close to or even exceed those of manual labor

Of course, not all data annotations can be replaced by AI. Some data annotations that require professional knowledge and logical analysis capabilities still require manual participation, such as medical, financial, autonomous driving and other fields. But this also means that the threshold of the industry will continue to increase. For data annotators, if they want to survive in this industry, they may need more learning and effort.

The above is the detailed content of The AI ​​large model data indicates that the monthly income of "migrant workers" is no more than 5,000 yuan, and the unit price has dropped from 5 cents to 4 cents.. For more information, please follow other related articles on the PHP Chinese website!