Bybit launches the first centralized-exchange live AI vs. human trading competition

Bybit announced a season-style contest that will pit externally developed algorithmic teams against human traders using the exchange’s live market infrastructure. Selected institutional and qualified participants must meet an entry capital threshold and provide technical documentation — including model descriptions and code repositories — so organizers can assess system behavior before granting API access. The contest enforces operational rules such as a required minimum number of trades per day and recommends leverage limits to reduce extreme risk-taking, while Bybit retains the right to disqualify entrants that breach compliance standards. Recruitment runs through early February, with a warm-up phase in mid-February and the competitive season slated to begin in March; final participant selection follows an internal screening process. Real-time profit-and-loss leaderboards and daily highlights are promised, creating a transparent performance trail for observers and potential institutional partners. By inviting academic, corporate, and research teams, the format seeks both innovation and measurement: it creates a controlled environment to compare machine-driven strategies with human decision-making under identical market exposures. From a technical perspective, the requirement to disclose interaction frequency, model architecture, and trading logic raises the bar for auditability but also exposes proprietary designs, which may deter some teams or require careful IP handling. Operational risks include latency differences, API rate limits, and the potential for automated strategies to exacerbate volatility; Bybit’s ongoing monitoring and screening aim to mitigate those concerns but do not eliminate them. The event could accelerate institutional interest in AI trading by providing a live, observable dataset on relative performance, which in turn may influence both product development and capital allocation decisions across quant trading desks. Regulators and compliance officers will likely scrutinize the contest for market integrity issues, including front-running, spoofing-like behaviors, and proper handling of customer protections when production infrastructure is used for competitive testing. For Bybit, the competition serves as both a marketing signal and a technical experiment: it demonstrates platform scalability for algorithmic participants while seeking to position the exchange at the center of AI-driven innovation in crypto markets. If successful and well-audited, the experiment may catalyze new vendor relationships and recruitment of institutional liquidity; if not, it could prompt calls for stricter exchange-level controls or third-party oversight. Overall, the initiative is a practical step toward normalizing AI in live trading, but its long-term impact will depend on transparency of outcomes, robustness of compliance procedures, and the community’s confidence in fair execution.