Coinbase fine-tuned an open-source AI model (Qwen3.5-9B) on its proprietary transaction data to enhance fraud detection. The experiment conducted on Onramp, which allows users to purchase cryptocurrencies through partner applications, resulted in a 30% reduction in fraudulent transactions and a 22% decrease in their value. The model was tested across 16,140 transactions involving 7,293 users, including 813 confirmed fraudulent transactions. CEO Brian Armstrong stated that specialized models outperform general-purpose models on specific tasks, at lower cost and with reduced latency.
Source: Read the original article
Disclaimer: this content is for information purposes only and is not financial advice. Cryptocurrencies are highly volatile: you may lose all of your capital. Always do your own research. Legal notice
