Grok and DeepSeek outperformed numerous prominent artificial intelligence chatbots in cryptocurrency trading, accurately timing the market’s local bottom ahead of a recovery rally, indicating a potential advantage for users who utilize their insights.
In a crypto trading competition that gained viral attention, Grok 4 and DeepSeek emerged as the top two generative AI chatbots.
DeepSeek registered a total unrealized profit of $3,650, becoming the most profitable chatbot, while Grok followed closely with around $3,000 in unrealized profits, as reported by blockchain data platform CoinGlass.
Claude’s Sonet 4.5 secured the third spot with $2,340 in generated profit, followed by Qwen3 Max, which managed $784 since the competition commenced.
Not all chatbots could secure profits; OpenAI’s ChatGPT 5 faced an unrealized loss of roughly $2,800, while Google’s Gemini 2.5 Pro recorded $3,270 in unrealized losses at the time of this writing.
Related: Amazon AWS outage disrupts Coinbase mobile app, Robinhood affected
AI chatbots like Grok and ChatGPT are becoming increasingly popular among crypto traders, aiding in the anticipation of potential altcoin rallies and identifying optimal entry points to avoid falling into exit liquidity.
While AI tools can assist in detecting shifts in real-time investor sentiment for day traders through social media and technical indicators, traders still cannot depend solely on them for autonomous trading.
Related: Elon Musk describes Bitcoin as energy-based and inflation-resistant, unlike ‘fake fiat’
Grok 4 anticipates crypto market bottom, shifts shorts to longs for 500% gain
XAI’s Grok 4 successfully predicted the market’s local bottom just before the recovery rally commenced.
Utilizing these insights, Grok transitioned from a previous short position to a long investment, achieving a 500% portfolio gain within the first day of the competition, stated Jaz Azhang, founder and CEO of AI company Stealth, in an Oct. 11 X post.
The chatbot executed numerous leveraged long positions on the two leading cryptocurrencies, including a 20x leveraged long on XRP (XRP), a 15x leveraged long on Solana (SOL), and a 10x long position on Dogecoin (DOGE).
“Thanks — markets reward sharp timing and conviction. Grok4 spotted the reversal early, converting risk into reward,” wrote Grok in a response to Azhang on Oct. 11.
In contrast, ChatGPT and Gemini maintained their initial short positions after the market bottom, leading to losses, indicating that these two models are less reliable for cryptocurrency traders.
The competition initiated with $200 in starting capital for each bot before scaling up to $10,000 per model, with trades executed on decentralized exchange Hyperliquid.
Magazine: Crypto traders ‘fool themselves’ with price predictions — Peter Brandt
