$Sahara is launching today, having quickly secured listings on all major exchanges. It's currently at $0.1448 on-chain, which means the market's pre-TGE valuation is $1.45 billion. The initial circulation is about 20%, or around $290M market cap. Compared to recent projects with similar TGEs, $Sahara's valuation and market cap are among the highest—only $NXPC can compare.

This also reflects the capabilities of the $Sahara team ⬇️

1. Funds and resources: "Quick listings on all major exchanges" prove that this team is not lacking in funds and resources. At the same time, they took care of retail investors in the TGE issuance rule design by conducting an IDO, allowing retail investors to also profit.

2. Narrative control: $Sahara's IDO used USD1 for fundraising, effectively leveraging the Trump WLFI stablecoin narrative. Combined with retail investors' previous preference for USD1 narrative speculation, this further increased the market's valuation (the subsequent quick listings on major exchanges further boosted the market's valuation).

However, since we're discussing $Sahara, it's necessary to explain what the project itself does.

Overall, $Sahara is a comprehensive decentralized AI platform with products that support AI developers in deploying and training AI models. It also has a decentralized AI marketplace that supports the trading of abstract assets such as computing power, models, and datasets. They also created Knowledge Agent, which is aimed at toC users (basically, users can ask Knowledge Agent questions, and after reviewing a series of materials, it will summarize and provide an answer. It can understand knowledge, remember knowledge, and use this knowledge to perform operations).

However, in my opinion, these are not the main focus. $Sahara's core product is actually data labeling. This is also a section I often mentioned in previous tweets. @TensorplexLabs, which Yzi Labs invested in, is also a product in the same sector.

A simple understanding of the data labeling concept is: AI training requires data, but the data is mixed, with both high-quality and low-quality data. Data labeling is essentially filtering the data and classifying and summarizing it according to different data types.

The core of this type of work lies in incentives—if you don't provide compensation, users have no motivation to do it. Tensorplex's method is to provide incentives to users by registering a Bittensor subnet. $Sahara is more direct this time, using airdrop expectations before the token launch and directly using $SAHARA for incentives after the token launch.

Then we can directly foresee the future development path of the $Sahara ecosystem: $Sahara incentivizes users to label data through tokens —> dataset uploads —> B-end customers purchase data through the AI market and use it for AI training. If the project isn't a one-off (although I think a valuation of $1.5 billion is a bit expensive), $Sahara will need to increase the token price to attract more people to participate in data labeling, in order to provide B-end customers with a richer selection of datasets and earn revenue from it (operating it as a cash flow income platform). This is basically how it works.

This approach is likely the ultimate path for all Crypto AI infrastructure projects—providing services to Web2 AI giants at a lower price through token-incentivized crowdsourcing (Web3 crowdsourcing). The other major category of Crypto AI projects, AI Agent, is another Ponzi scheme.

Of course, $Sahara is a large platform, and there are many product lines that can be developed in the future besides data labeling. This depends on the team's own ideas and execution.

All You Need to Know in 10s

TermsPrivacy PolicyWhitePaperOfficial VerificationCookieBlog