MEXC Survey: 85% of Crypto Traders Use AI Despite Restricting Fund Control

Mutsamudu, Comoros, October 16, 2026 — Global cryptocurrency exchange MEXC has published its AI Trading Report, drawing insights from a survey of 3,708 digital asset traders. The study investigates how participants utilize artificial intelligence, the extent of trading authority they are willing to grant, and the safeguards they demand from exchanges. Blockchain security auditor Hacken also contributed technical analysis examining the risks tied to AI agent trading alongside the controls built to mitigate them.

Key takeaways:

• 84.8% of surveyed crypto users have employed AI for research or trading. While 58.3% cite enhanced market research, 46% note better trade execution, though 54% have yet to experience improvements at the execution stage.
• Interest in AI agents is substantial, with half of respondents expressing interest despite not yet having tried them.
• 65.9% of AI users act on its guidance at least on occasion, while only 20% have tested execution without manually approving each transaction.
• 54% are open to letting AI trade with a portion of their funds, though capped at a maximum of 20% of their account balance; merely 6.5% would grant access to over half.
• Half of users with automated execution experience worry they might fail to spot or halt problems in time.
• Willingness to utilize automated trading climbs from a baseline of 55% to 71.8% when restrictions on capital, positions, leverage, and losses are implemented.
• 95.5% endorse at least one mandatory security safeguard, even as many advocate for personal control over their custom settings.

The study revealed that 84.8% of participants had utilized AI to evaluate or trade digital assets over the preceding six months, with 53.1% engaging with it at least weekly. Younger demographics demonstrated higher adoption rates: 88.5% of users aged 18–34 utilized AI, compared to 69.9% of those 45 and older. Among non-users, complications getting started or high entry barriers proved to be more prevalent hurdles than a general mistrust of AI, cited by 46.7% and 32.7% of respondents, respectively.

Pre-trade preparation remained AI’s primary function. Summarizing news stood out as a top task for 54.3% of AI users, trailing market analysis at 48.6% and discovering trading opportunities at 38%. Only 7.6% identified direct execution as a primary task. Reflecting on their most recent AI-assisted transaction, 68.9% noted that AI contributed most significantly prior to trade execution, versus 22.7% during the actual trading phase.

Respondents generally held more favorable views toward research than execution. While 58.3% reported superior market research, 46% observed improved execution. Conversely, 30.9% noted no shift in execution quality, and 23.1% stated it deteriorated. These evaluations stem from user experiences rather than independently quantified trading performance metrics.

Nonetheless, AI recommendations directly impacted trading choices. Among AI users, 65.9% acknowledged executing orders or modifying positions based on its suggestions at least sometimes, acknowledging that these categories could overlap.

The presence of risk parameters significantly influenced automation willingness. Under a baseline scenario offering notifications, pausing, and permission revocations, 55% were comfortable letting AI trade autonomously. Incorporating limits on funds, assets, positions, leverage, and maximum losses—with automated shutdowns triggered at loss thresholds—elevated willingness to 71.8%, marking a 16.8 percentage point increase.

However, that openness did not extend to managing the bulk of an account’s capital. In a separate scenario that barred transfers and withdrawals while enabling manual halts at any time, 54% stated they would risk no more than 20% of their account funds, and just 6.5% would entrust over half.

Hacken’s analysis highlights why restrictions on withdrawals alone are insufficient to block financial losses. An agent with trading privileges could still be manipulated into acquiring an asset at inflated valuations or assuming excessive exposure. Their assessment highlights the necessity of protective barriers configured independently of the agent—such as restrictions on orders, aggregate positions, and permissions—alongside security validations to intercept unauthorized actions prior to execution.

Regarding platform governance, 46.2% of participants favored exchanges supplying safety configurations that users could selectively toggle, roughly 1.8 times the 25.7% who preferred uniform baseline mandates established by platforms. This preference for personal customization did not translate to a rejection of baseline standards: among respondents favoring optional settings who also addressed the standards question, 97.8% backed a minimum of one mandatory safeguard.

Across all 3,464 valid responses regarding regulatory standards, risk evaluations or literacy tests prior to initial use drew the highest backing at 55.3%. Independent security audits and mandatory caps on capital and losses that cannot be turned off followed closely at 38.8% each.

“As AI becomes more integrated into trading, our findings show that users don’t want less control. They want clearer boundaries,” said Vugar Usi Zade, CEO of MEXC. “For traders, that means clear risk limits, the ability to intervene when needed, and confidence that the platform has done its part before automation begins.”

The complete AI Trading Report, including Hacken’s security evaluation, can be accessed here.

About MEXC

Founded in 2018, MEXC is a leading global multi-asset trading platform built as your 0-fee gateway to infinite opportunities. Serving users across 170+ markets, MEXC provides simple and efficient access to crypto, stocks, tokenized assets, derivatives, and a growing range of TradFi-linked opportunities through one account and one gateway.

With 0 trading fees, deep liquidity, broad asset coverage, and a high-performance trading experience, MEXC is designed for retail users who want to discover earlier, act faster, and trade with fewer barriers. As crypto and traditional finance continue to converge, MEXC is committed to making global opportunities more accessible, helping users trade freely and MEXCmize every opportunity.

MEXC Official Website｜ X ｜ Telegram ｜How to Sign Up on MEXC

For media inquiries, please contact the MEXC PR team: media@mexc.com

Source

Short FAQs

• What percentage of crypto traders have used AI according to the MEXC report?
  The report found that 84.8% of surveyed crypto users have used AI for research or trading.
• How much of their account funds are traders willing to let AI manage?
  54% of respondents would let AI trade with some funds, but no more than 20% of their account balance, with only 6.5% willing to authorize more than half.
• Does adding risk limits increase users’ willingness to use automated trading?
  Yes, willingness grows from 55% under a baseline scenario to 71.8% when limits on funds, positions, leverage, and losses are introduced.
• What security firm contributed to the report?
  Blockchain security firm Hacken contributed technical analysis regarding the risks and controls associated with AI agent trading.