Binance says AI-driven security foiled $10B in fraud since 2025.

Binance says its AI-powered security tools prevented more than $10 billion in user fund losses to scams and fraud between 2025 and March 2026, according to a security-focused blog post from the exchange. The company reported that over 5.4 million users were protected from fraud between Q1 2025 and Q1 2026 after deploying more than 24 AI-driven initiatives and more than 100 models.

Over the 15 months to March 2026, Binance says it blocked $10.53 billion of potential losses and blacklisted 36,000 malicious addresses as its AI-enabled security protocols ran in the background. In the most recent quarter, the exchange highlighted a surge in activity: it intercepted 22.9 million scam and phishing attempts in Q1 2026 alone, saving $1.98 billion for users. The numbers, described in Binance’s security blog, underscore the rapid expansion of automated defenses as scams grow more sophisticated.

“AI-powered scams and exploits are accelerating,” Binance said. “The barrier to entry for scam perpetrators is falling fast, with AI accelerating the drop. What once required technical expertise can now be executed for next to nothing and at scale.” The warning reflects broader industry concerns that threat actors are leveraging artificial intelligence to enhance impersonation, social engineering and phishing campaigns. In the same vein, the FBI has noted elevated scam activity tied to crypto, citing substantial losses reported in the United States in recent periods.

Binance emphasizes that its AI tools are not just reactive, but embedded in real-time decisioning across its security stack. By integrating AI with its risk controls, the platform says it now intercepts attacks before user funds are moved and helps to differentiate legitimate activity from fraudulent patterns at scale. According to the blog, AI-driven decisioning now powers 57% of its fraud controls, contributing to a 60-70% reduction in card fraud rates relative to industry benchmarks. The company also pointed to ongoing work in identity verification to counter increasingly sophisticated deepfakes and synthetic identities.

Key takeaways

• Over 15 months ending in March 2026, Binance reports $10.53 billion in prevented losses and 36,000 malicious addresses blacklisted through AI-enhanced security measures, with more than 5.4 million users protected between Q1 2025 and Q1 2026.
• In Q1 2026 alone, the exchange intercepted 22.9 million scam and phishing attempts, saving $1.98 billion for users.
• AI-driven decisioning accounts for 57% of Binance’s fraud controls, with a claimed 60-70% reduction in card fraud rates versus industry benchmarks.
• The security stack employs computer vision to verify payment proofs, real-time language analysis to detect scam patterns, and enhanced identity verification to combat deepfakes and synthetic identities.
• The development comes amid a broader concern that AI-enabled scams are increasing in breadth and sophistication, reinforcing the need for robust defense mechanisms across exchanges and custodial platforms.

Inside Binance’s AI security toolkit

The security program described by Binance folds a wide range of AI capabilities into its existing risk controls. The company notes that computer vision components are used to detect fake payment proofs, while real-time language analysis monitors communications for suspicious patterns and indicators of fraud. On the identity side, the system is designed to counter deepfakes and synthetic identities that fraudsters may use to bypass traditional verification steps. Taken together, these tools are meant to reduce the likelihood that compromised or impersonated accounts can be exploited for exfiltration or manipulation of funds.

Binance also highlights a data-driven approach to governance and monitoring. By combining behavioral signals, transaction patterns and contextual data, the platform aims to flag unusual activity early and apply automated responses in real time. The emphasis on AI-driven decisioning—where a majority of fraud controls are now powered by AI—speaks to a broader shift in crypto security toward automated, scalable defense mechanisms that can adapt to evolving attack tactics.

Context, risk, and regulatory undertones

The concerns Binance raises about AI-enabled scams echo a broader industry risk landscape. Crypto-related fraud remains a persistent challenge for users and regulators alike, with authorities noting that attackers rely on social engineering, impersonation and increasingly convincing fake content. The FBI has flagged substantial losses linked to crypto scams, underscoring the importance of effective defense mechanisms as the technology that underpins these platforms scales in capability and reach. These dynamics place exchanges under pressure to demonstrate tangible security improvements while maintaining a user-friendly experience and transparent reporting.

From a market perspective, the deployment of AI in security can influence user confidence, willingness to engage with centralized exchanges, and the perceived risk profile of asset custodians. Investors and traders will be watching not only the absolute totals of funds saved or losses prevented but also how these tools perform across different jurisdictions, regulatory environments and product segments. The question going forward is how AI-driven security measures will balance proactive defense with user privacy, provide explainable risk signals, and scale with the growing volume and variety of crypto activity.

What to watch next

Readers should monitor whether Binance expands its AI security footprint beyond its current scope, including coverage of new product areas, regions, or asset classes. While the firm cites substantial quarter-to-quarter improvements, questions remain about the durability of these defenses as attackers adapt and as AI itself evolves. Expect more granular updates on the effectiveness of individual AI modules, potential cross-border enforcement considerations, and how other exchanges position their own AI-driven security programs in response to rising threats.

For a closer look at Binance’s security measures and the full set of claims, see the firm’s security blog post detailing the initiative and the metrics cited accompanying the 24 AI-driven initiatives and 100+ models. The post highlights the linkage between AI tools and real-time risk controls, emphasizing the intention to harden defenses as scammers grow more sophisticated. Binance security blog.

Alongside these developments, observers will continue to weigh the role of AI in fraud prevention against the risk of over-reliance on automated systems, potential false positives, and the need for ongoing user education. The crypto ecosystem remains a dynamic arena where defensive innovation and adversarial ingenuity co-evolve, and Binance’s latest figures offer a data point in that ongoing confrontation.

As the sector digests these disclosures, the key takeaway for users and investors is clear: AI-assisted security is becoming mainstream in crypto, but it is not a panacea. Ongoing transparency about performance, edge cases, and the safeguards around automated decisioning will be essential to gauge whether these tools translate into reliably safer experiences for the average user. The next few quarters should reveal whether the early momentum translates into sustained reduction in fraud across the broader ecosystem.

This article was originally published as Binance says AI-driven security foiled $10B in fraud since 2025. on Crypto Breaking News – your trusted source for crypto news, Bitcoin news, and blockchain updates.