2026-01-15 13:56:18

Aster "Human vs AI" Trading Duel Results Are Out. Surprised? Human trader ProMint really stands out — one person managed to beat the entire team. But the overall picture is a bit heartbreaking: 43% of traders got liquidated, and the team's ROI dropped to -32.22%. In contrast, among 30 AIs, none got liquidated. Although the overall ROI is also negative (-4.48%), their stability is in a completely different league.

Numbers say everything. Among human traders, individual differences are huge — some can profit, while most are stuck in a losing streak. AIs are different; they seem to be equipped with stabilizers. Although they are also declining, their risk defense line is firmly maintained. This is not to say AI is invincible, but in volatile markets, risk control awareness and execution are obviously more composed in machines.

Interestingly, this duel actually points to a direction: the future is not about humans being replaced by AI, but how to collaborate. Human intuition and market sentiment sensing have unique value; AI’s discipline and risk management complement human weaknesses. The next showdown is scheduled for January 22, and new stories will unfold then.

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

12 Likes

Reward
12
6
Repost
Share

Comment

0/400

SerRugResistant

· 5h ago

43% liquidation haha, this is the daily life of retail investors, AI is really stable to an incredible degree. ProMint's guy is really impressive, but he's probably just a survivor bias. Huh, not a single liquidation with AI? That requires a lot of caution. Come back on January 22nd, I bet AI will still win. Human-machine collaboration sounds good, but can it really be implemented effectively? Risk control is just risk control, AI is more sophisticated than humans. This data is a bit painful to face, how are those people in the group doing? -32 looks uncomfortable, better to play it safe with AI. Wait, is there still a chance for humans to turn things around? As long as the AI stabilizer has no bugs, I'm watching.

View OriginalReply0

FarmHopper

· 8h ago

43% liquidation is really outrageous; humans are still too greedy. AI stability is stable, but when the market really comes, can we seize the opportunity? That guy from ProMint is really awesome. How did he do it? Next year still depends on how people coordinate with AI; relying on anyone alone won't work. I have to hold onto the January 22nd showdown; will there be a reversal this time?

View OriginalReply0

AllInAlice

· 8h ago

ProMint that guy is really ruthless, carrying the entire team and dragging it down AI is really stable, although it also loses money but at least it’s alive Humans are still too emotional, machines are just tough 43% liquidation data is incredible, I just want to laugh when I see it Cooperation is the way to go, pure AI or pure humans both have a ceiling Next time on January 22nd, I also want to see if ProMint can still carry I believe that AI won’t get liquidated, the stability is indeed worlds apart The whole team -32%, AI only -4.48%, the gap is obvious Human intuition combined with AI discipline, this combination is truly excellent I love this kind of showdown, but it feels like the result is already obvious

View OriginalReply0

ProposalManiac

· 8h ago

This data structure is quite clear—essentially an incentive compatibility problem. On the human side, high returns come with high risks; a 43% liquidation rate essentially means the risk control mechanism has failed. AI is more stable but has an obvious收益 ceiling; this is not about who is stronger, but about two completely different game-theoretic logics. What truly matters is how to design this human-machine collaboration framework moving forward. Simply letting AI守风险下线 (manage risk offline) and humans负责alpha (responsible for alpha) requires clarifying this权力分配 (power distribution)—who has the final decision-making authority? How are the收益 (returns) split? These governance details determine how far the collaboration model can go. Historically, even top-tier mathematician teams like LTCM eventually failed, and the key factor is human nature. It’s easy to装上风控枷锁 (put risk control shackles) on AI, but the难 (difficult) part is to make humans truly trust the机制 (mechanism) and not override it at critical moments. As for the showdown on January 22, I’m curious whether ProMint can replicate its success or if that was just an outlier.

View OriginalReply0

LiquidityWitch

· 8h ago

ProMint single-handedly took over the entire scene, hilarious. Is this what humanity has come to? AI stability is indeed impressive, but losses are losses. Nothing to boast about. Collaboration is the real key; relying solely on machines is too boring. The 43% liquidation rate is heartbreaking; I almost became one of them. Waiting for the 22nd to see if there will be a reversal.

View OriginalReply0

BTCWaveRider

· 8h ago

That's enough, it's the same old story. When AI wins, everyone praises AI; when humans win next time, they'll praise human intuition. The real issue is, what about those guys who got liquidated with 43% loss...

View OriginalReply0