TAG (TAG) Price

Tagger is a platform covering AI data collection, labeling, management, and trading; TAG is the token used by the platform to incentivize qualified data contributions. Data workers receive TAG only after their submitted data passes validation by the platform's AI system. TAG rewards are therefore different from proof of ownership of a dataset and do not arise automatically from merely submitting data. In the data collection process, after a task publisher initiates a task, the platform uses natural language processing technology to analyze the task's semantics and match relevant data categories, then connects the task with data workers. Workers upload their contributions, and only after platform validation is the amount of TAG determined based on the data's value and quality.

Participants do not receive TAG simply by uploading; after a data contribution passes the platform's AI validation, they receive a specified amount of TAG according to that data's value and quality. The validation step distinguishes acceptable data contributions from submissions that do not pass validation, so the act of submitting itself does not guarantee a reward. In the process, a task publisher first posts a data collection task, and the platform analyzes the task's semantics and matches data categories. Data workers then upload data accordingly, and the AI system validates their contribution; reward settlement begins only after it passes. The reward amount is based on the value and quality of validated data, so the number of tasks cannot be directly treated as the final TAG received.

For AI Copilot labeling tasks, TAG rewards are the product of four factors: the per-task reward, halving coefficient, account-level coefficient, and mask-accuracy coefficient. In other words, with other conditions unchanged, mask accuracy directly changes the reward for that task through its coefficient, rather than serving only as a general quality assessment. The formula can be written as: TAG reward = per-task reward × halving coefficient × account-level coefficient × mask-accuracy coefficient. The official mask-accuracy coefficients are: 0 when accuracy is below 85%; 0.4 from 85% to below 90%; 0.8 from 90% to below 95%; 1.0 from 95% to below 97%; and 1.5 at 97% and above. Therefore, the multiplier is zero below 85%, while reaching a higher accuracy range increases this factor's effect on the final reward.

After data collection is completed, the task publisher can generate a unique NFT as proof of its rights to the dataset. This NFT confirms dataset rights; it is not a replacement for the TAG reward received by data workers after validation, as the two concern different subjects and functions in the process. Specifically, data submitted by workers that passes validation triggers TAG reward settlement. Once the data collection process is complete, the task publisher generates an NFT to assert rights to the dataset. Thus, receiving TAG reflects an incentive for qualified contributions, while the NFT reflects confirmation of the task publisher's rights to the completed dataset.

The Tagger marketplace allows data owners to choose to sell datasets or license their data to third parties; the two methods correspond to transfer and licensed use, respectively, and are not the same thing. Transfer and transaction records related to on-chain data-rights credentials are recorded on the blockchain to trace how rights move. Therefore, after an NFT confirms dataset rights, the marketplace layer handles how the dataset associated with that rights credential is traded or licensed. On-chain records can provide traceability for transactions and transfers, but whether to sell, whom to license to, and the specific scope of use still depend on the relevant marketplace transaction or licensing arrangement.