Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
According to the ChipMaker Nvidia, according to one of the Synthetic AI developers, according to the purchase of a synthetic information firm, according to a A. String report Wednesday.
Moving, generative AI firms, struggle to find enough information to cultivate models by increasing the need to create information and develop models.
According to the report, Gretel’s employees will be folded to NVIDIA. Gretel, which produces synthetic or simulated information for AI model education, will restore its proposals for NVIDIA’s AI developers.
A NVIDIA spokesman refused to comment on the report.
Why synthetic data is important
Openai’s Chatgpt, a large language model, generation AI models require a large number of information. Real-world data can create a problem for AI developers – that is, it can be noisy and not enough.
AI firms continue to the limit of training information that causes conflicts related to copyright rights. Hundreds of actors, writers and directors submitted an open letter To increase the concerns about the use of copyrighted data to the office of the Trump Administration of Science and Technology. Currently, the Openai government applies to purchase Allow more access to copyrighted material Raising AI models or other American companies will be left behind by China.
See that: See NVIDIA’s GTC 2025 Keynote: All points in 16 minutes
Synthetic data also has the value in protecting personal data. Gretel says synthetic information Can be used to develop models and means without exposing sensitive or personal information – for example, personal people do not recognize people and potentially violate the privacy laws.
Have Concerns about the use of such information In model training. According to information, a model of a brand can increase the possibility of incorrect. If the problem is bad enough, it can cause a problem known as the collapse of the model.
