News

Foresight News reported, according to monitoring by Lookonchain, Cooker.hl spent 6 SOL (approximately $867) to purchase 16.5 million PENGUIN, and then sold all of them for 6.12 SOL (approximately $886), making a profit of only 0.12 SOL (about $18). The 16.5 million PENGUIN are now worth $1.4 million, meaning this address missed out on $1.4 million in profit.

PANews, January 24 — According to on-chain data monitoring, in the past hour, a certain address has continued to buy PENGUIN at high levels, accumulating a position of $260,000. Notably, this address had previously made a precise purchase of WHITEWHALE near its local low, demonstrating a certain "contrarian timing" record. Meanwhile, another address has adopted the opposite strategy, having completed its main accumulation of PENGUIN at lower prices and is now taking profits in batches as the price rises. This address still holds PENGUIN worth over $600,000, with a paper return of approximately 36,000%. Analysis suggests that within the current PENGUIN price range, there is a clear divergence in capital behavior, with both high-level chasing and long-term position realization coexisting.

Jinse Finance reported that a16z Crypto published an article titled "How AI 'Judges' Can Scale Prediction Markets," which pointed out that the biggest challenge prediction markets face is not "pricing the future," but determining what actually happened. Similar issues frequently arise in some small-scale events, where incorrect or non-transparent settlement mechanisms can undermine market trust, liquidity, and the accuracy of price signals. Industry experts suggest introducing large language models (LLMs) as arbiters in prediction markets, including on-chain rule commitments, resistance to manipulation, enhanced transparency, and increased neutrality. For example, when creating a contract, the specific LLM model, timestamp, and judgment prompt can be encrypted and recorded on the blockchain, allowing traders to understand the complete decision-making mechanism in advance. The model's weights are fixed and cannot be easily tampered with, reducing the risk of cheating. The settlement mechanism is open and auditable, with no arbitrary human rulings. The AI judgment mechanism can significantly improve the efficiency and scalability of prediction market settlements while ensuring transparency and fairness.