Binance says its AI systems stopped $10.5 billion in fraud since 2025 as crypto scam costs collapse

Binance says its AI-powered security systems have blocked more than $10.5 billion in potential fraud losses since early 2025, with the exchange intercepting nearly 23 million phishing attempts in the first quarter of 2026 alone, as the cost of launching AI-assisted crypto attacks falls to as little as $1.22 per contract.

This avoided attack protected nearly $2 billion in user assets during that period. The exchange said more than 100 AI models are now embedded across its risk management and fraud prevention systems.

The announcement comes at a time when crypto-related fraud is evolving from phishing schemes into AI-enhanced attacks involving deepfakes, voice cloning, automated impersonation bots, and social engineering tactics.

AI-powered scams are reshaping crypto security

Binance’s disclosure reflects a broader trend across the digital asset industry, where artificial intelligence is becoming both a weapon for cybercriminals and a defensive tool for exchanges.

According to blockchain intelligence firm TRM Labs, AI-enabled crypto scams surged by 500% in 2025, with fraudsters increasingly using generative AI to automate phishing campaigns and impersonation attacks.

The exchange said AI-driven fraud controls now account for roughly 57% of its security infrastructure. Binance also claimed its systems reduced card fraud rates by 60% to 70% below broader industry benchmarks.

The company attributed much of the recent threat escalation to the falling cost of AI-powered exploits.

Binance Research estimated that AI-assisted smart contract attacks can now cost as little as $1.22 per contract, significantly lowering the barrier to entry for malicious actors.

Industry analysts say this trend is forcing exchanges to shift from reactive security models toward predictive threat detection systems capable of identifying suspicious patterns before funds leave user accounts.

Exchanges race to strengthen fraud prevention systems

Binance is not the only exchange investing heavily in AI security infrastructure.

Earlier this year, Bybit said its AI-assisted risk monitoring system prevented more than $300 million in suspected scam-related withdrawals during the fourth quarter of 2025.

The exchange stated that real-time alerts and automated withdrawal screening protected more than 4,000 users from potentially fraudulent transfers.

AML and blockchain analytics firms are also warning that social engineering attacks have overtaken smart contract exploits as the dominant threat facing retail crypto users.

According to blockchain security firm AMLBot, approximately 65% of crypto-related incidents investigated in 2025 involved impersonation schemes, phishing attacks, and compromised devices.

That shift is particularly significant for investors, because even well-audited blockchain systems remain vulnerable when attackers exploit human behavior instead of code weaknesses.

Investor confidence remains tied to trust and compliance

For Binance, the report also serves as a reputational milestone as the exchange continues rebuilding trust after years of regulatory scrutiny.

The company has faced ongoing criticism over anti-money laundering compliance and transaction monitoring, including allegations reported by the Financial Times.

Against that backdrop, Binance’s latest AI security disclosures appear aimed at reassuring both regulators and institutional investors that the exchange is strengthening internal safeguards as crypto adoption expands globally.

The company said it reviewed more than 71,000 law enforcement requests during 2025 and helped authorities confiscate approximately $131 million in illicit funds.

Binance also claimed it recovered $12.8 million across 48,000 fraud-related cases, marking a 41% year-over-year increase in recovery activity.

As AI-generated scams become cheaper, faster, and harder to detect, exchanges capable of deploying large-scale automated defenses may gain an advantage in retaining both retail and institutional users.

Still, security experts caution that platform-level protections alone are unlikely to eliminate fraud risks entirely.

The views expressed in content distributed by Newstex and its re-distributors (collectively, "Newstex Authoritative Content") are solely those of the respective author(s) and not necessarily the views of Newstex et al. It is provided as general information only on an "AS IS" basis, without warranties and conferring no rights, which should not be relied upon as professional advice. Newstex et al. make no claims, promises or guarantees regarding its accuracy or completeness, nor as to the quality of the opinions and commentary contained therein.