Skynet Hack3D Report Highlights Web3 Security in 2025

TheNewsCryptoPublished on 2025-12-24Last updated on 2025-12-24

Abstract

Web3 entered 2025 with strong momentum due to improved macroeconomic conditions, supportive U.S. policies, and growing institutional confidence. However, total losses from cyberattacks reached $3.35 billion, a 37% increase from 2024. Excluding the $1.45 billion Bybit exploit—attributed to the Lazarus Group via a supply-chain attack—losses actually decreased year-over-year, indicating a shift toward fewer but more sophisticated and high-value attacks. Phishing was the most common attack vector, accounting for $722.9 million across 248 incidents, followed by code-related exploits at $554.6 million. AI played a dual role: enhancing defensive tools like automated audits while also enabling highly convincing phishing campaigns, deepfake impersonations, and rapid exploit replication. Regulatory clarity improved with the U.S. GENIUS Act and E.U.’s MiCA implementation, raising security and compliance standards. CertiK expanded its security offerings, integrating risk analysis tools, publishing major reports, and advancing research in zkEVM and consensus mechanisms. As Web3 grows, integrating security at all levels remains essential to mitigate evolving AI-driven and systemic risks.

The Web3 ecosystem entered 2025 with renewed momentum, buoyed by improving macroeconomic conditions, stronger investor confidence, and a noticeably more supportive political climate in the United States. The new U.S. administration moved quickly to position digital assets as a strategic innovation sector rather than a regulatory anomaly, sending an early signal that blockchain technology would be encouraged rather than restrained. This shift restored confidence among builders, institutions, and venture capital, helping decentralized applications expand deeper into payments, gaming, tokenized assets, identity solutions, and real-world financial use cases.

Yet, as activity accelerated across the ecosystem, so did the threat landscape. Cyber adversaries evolved alongside the industry, refining both technical exploits and social engineering techniques. While innovation surged, 2025 became a stark reminder that growth and risk continue to move in parallel within Web3.

According to industry data, total losses in 2025 reached $3.35 billion, marking a 37% increase compared to $2.45 billion in 2024. At first glance, the numbers suggest a dramatic deterioration in security conditions. However, a closer look reveals a more nuanced picture. One single incident, the Bybit exploit, accounted for approximately $1.45 billion of the year’s losses. When this outlier is excluded, overall stolen funds would have declined year over year, underscoring a critical shift in attacker behavior.

Rather than relying on a high volume of mid-sized exploits, threat actors increasingly concentrated resources into fewer but far more devastating operations. The Bybit incident demonstrated the growing presence of well-funded, highly coordinated adversaries capable of executing complex, long-horizon attacks. This trend suggests that while baseline security hygiene is improving across many protocols, systemic risks remain, particularly at the infrastructure and supply-chain level.

When categorizing attack vectors, phishing emerged as the most prevalent threat in 2025. Excluding the Bybit supply-chain breach, phishing accounted for $722.9 million stolen across 248 incidents, surpassing both code vulnerabilities and infrastructure attacks in frequency. Code-related exploits followed closely, resulting in $554.6 million across 240 incidents, although nearly half of those funds were eventually frozen or returned, highlighting improved response coordination and on-chain intervention capabilities.

Artificial intelligence played a defining role in shaping this evolving threat environment. On the defensive side, developers increasingly relied on AI-powered tools to generate test cases, identify inefficiencies, enhance formal verification, and streamline audit workflows. Conversely, attackers adopted the same technologies at scale. AI-generated phishing interfaces became nearly indistinguishable from legitimate dApps and wallet prompts, while automated multilingual campaigns expanded reach into previously insulated communities.

Threat actors also leveraged AI for reconnaissance, scraping on-chain data and private chat channels to identify high-value targets. Impersonation attacks grew more convincing, with fake founder accounts, synthetic voices, and deepfake videos eroding traditional trust signals. Perhaps most concerning was the speed of exploit replication, as AI tools enabled attackers to copy and deploy successful attack patterns within days or even hours.

Regulatory clarity improved significantly throughout 2025, helping stabilize the broader ecosystem. In the U.S., the GENIUS Act established early frameworks for stablecoin oversight and digital asset transparency, while signaling a more cooperative stance toward innovation. Globally, the European Union advanced toward full MiCA implementation, raising standards for disclosures and consumer protection. Meanwhile, jurisdictions such as Singapore and Hong Kong expanded digital asset sandboxes, and countries including Brazil and Colombia progressed toward regulated commodity tokenization frameworks.

These developments contributed to more structured governance and influenced how projects approached compliance, architecture, and operational security. As regulations matured, security increasingly became a prerequisite for market access rather than an optional feature.

One of the year’s most significant incidents occurred in February, when Bybit suffered the largest crypto theft in history. The attack, attributed to the Lazarus Group, did not exploit Bybit’s internal systems directly. Instead, attackers compromised a developer machine at Safe{Wallet}, a third-party multi-signature wallet provider. Malicious code injected into the wallet interface altered transaction details invisibly, causing authorized signers to unknowingly approve fraudulent transfers. The incident exposed the growing risks associated with trusted tooling and supply-chain dependencies.

Beyond large-scale breaches, individual users faced mounting risks. AI-driven phishing, deepfake impersonation, and targeted social engineering attacks surged throughout the year. Many losses went unreported, particularly those linked to off-chain scams such as pig-butchering schemes and investment fraud, suggesting that actual user losses are likely far higher than recorded figures.

As 2026 approaches, the trajectory of Web3 security is becoming clearer. Attackers are expected to further refine AI-powered impersonation and phishing campaigns, while supply-chain attacks may grow more sophisticated. At the same time, stronger regulation, real-time monitoring, and AI-assisted defenses offer a path toward reducing preventable losses.

2025 at CertiK

2025 marked a milestone year for CertiK, defined by expanded research, deeper ecosystem integrations, and continued leadership in Web3 security. Below are some of the key achievements that shaped the year:

  • Integrated Token Scan with ChainGPT and Binance Wallet, extending real-time token risk analysis directly into widely used Web3 tools.
  • Published the Skynet Stablecoin Spotlight Report: H1 2025, delivering an in-depth review of the stablecoin landscape, key vulnerabilities, and how the Skynet Security Score can be used to assess stablecoin risk.
  • Released the 2025 Skynet RWA Security Report, providing structured due-diligence criteria and a comprehensive risk review framework for real-world asset (RWA) protocols.
  • Launched the 2025 Skynet Korea Web3 Security & Ecosystem Report, offering insights into South Korea’s Web3 market dynamics and profiling leading platforms in the region.
  • Published the 2025 Skynet Digital Asset Treasuries (DAT) Report, introducing the Skynet DAT Security & Compliance Framework to evaluate operational integrity beyond surface-level metrics.
  • Released the Skynet U.S. Digital Asset Policy Report, summarizing the legal foundations, market-structure implications, and operational requirements of the GENIUS Act and CLARITY Act in the United States.
  • Conducted a full-scale security assessment of the USDCx mint and burn process on Canton Network, including audits of on-chain Daml smart contracts and penetration testing of off-chain infrastructure.
  • Launched CertiK SkyNode, a validator node service designed to improve network security, reliability, and performance across multiple public blockchain ecosystems.
  • Co-published research with Ant Group’s AntChain (Ant Dense Computing) focused on the formal verification of core components within the Asterinas operating system.
  • Introduced the LiDO framework, presented by CertiK Co-Founder Professor Shao Zhong, addressing critical security challenges in Byzantine Fault Tolerant (BFT) consensus mechanisms.
  • Secured two grants from the Ethereum Foundation, reinforcing CertiK’s leadership position in zkEVM formal verification research.
  • Rolled out Skynet leaderboards, a security-focused ranking platform designed to evaluate and compare crypto and Web3 project security.
  • Released ecosystem-specific showcase leaderboards to support strategic Layer 1 growth, including dedicated leaderboards for BNB Chain and SUI..

In this rapidly evolving environment, long-term success will depend on integrating security into every layer of Web3 development. As the largest Web3 security services provider, CertiK continues to play a central role in safeguarding the ecosystem, supporting thousands of projects, and strengthening trust as blockchain technology moves closer to mainstream adoption.

TagsBlockchainexchange

Related Questions

QWhat was the total value of losses in the Web3 ecosystem in 2025, and how did it compare to the previous year?

ATotal losses in 2025 reached $3.35 billion, which was a 37% increase compared to the $2.45 billion lost in 2024.

QWhich single incident was responsible for the largest portion of losses in 2025, and how much was stolen?

AThe Bybit exploit was the largest single incident, accounting for approximately $1.45 billion of the year's total losses.

QWhat was the most prevalent type of attack vector in 2025, excluding the Bybit incident?

APhishing was the most prevalent attack vector, accounting for $722.9 million stolen across 248 incidents.

QHow did artificial intelligence (AI) impact the Web3 security landscape in 2025?

AAI was used both defensively by developers for tasks like generating test cases and enhancing audits, and offensively by attackers for creating convincing phishing interfaces, automated multilingual campaigns, reconnaissance, and rapid exploit replication.

QWhat significant U.S. regulatory development for digital assets is mentioned in the report?

AThe GENIUS Act established early frameworks for stablecoin oversight and digital asset transparency, signaling a more cooperative U.S. stance toward innovation.

Related Reads

From South Korea to the United States: Blue-Collar Jobs Are Becoming Increasingly Popular, Thanks to AI

AI is reshaping the labor market's value proposition. The traditional four-year college degree is losing its appeal as a guaranteed career path, while skilled blue-collar trades like electricians, welders, and plumbers are experiencing historic demand and wage premiums. This shift is driven by dual pressures: AI's displacement of certain white-collar roles and a booming need for physical infrastructure and data center construction. Data confirms the trend. In the U.S., vocational school revenue surged, and a significant portion of recent layoffs are AI-related. Surveys show a majority of Gen Z adults plan to pursue blue-collar work, citing better job security against AI automation. Vocational education interest has exploded recently. Experts cite a psychological shift as younger generations seek tangible, AI-resistant careers and avoid high student debt. In many cases, salaries for skilled trades now match or exceed those requiring a bachelor's degree. In South Korea, semiconductor vocational high schools boast near-total employment, with graduates securing high-paying roles at companies like Samsung. The shortage is structural, exacerbated by a retiring baby boomer workforce and massive infrastructure projects. Companies like JPMorgan Chase, Meta, and Lowe's are investing heavily in training programs. However, overcoming historical stigma and a "perception gap" around trade careers remains a key challenge to closing the talent gap.

marsbit1h ago

From South Korea to the United States: Blue-Collar Jobs Are Becoming Increasingly Popular, Thanks to AI

marsbit1h ago

Qualcomm: AI Hype Subsides, When Will Smartphones Emerge from the Gloom?

Qualcomm reported its Q3 FY2026 results (ending June 2026), with revenue of $9.95B, down 4% YoY but above expectations. Gross margin declined to 53.1%, pressured by rising costs across manufacturing and memory. Key business segments showed mixed performance: Handset revenue fell 19.6% YoY to $5.09B, dragged by an 11% decline in non-Apple Android shipments and weaker high-end mix. Conversely, Automotive revenue surged 61% to $1.59B, and IoT grew 9% to $1.83B. Core operating profit dropped 41% YoY due to margin compression and higher expenses. Management's Q4 FY2026 guidance projects revenue of $9.7B-$10.5B, in line with consensus, but Non-GAAP EPS guidance of $2.05-$2.25 fell short of expectations. Amidst persistent weakness in its core handset market, Qualcomm is pursuing growth in AI, focusing on Edge AI (smartphones, PCs, automotive) and Data Center AI. Its data center strategy includes four pillars: AI accelerators (e.g., AI200), commercial CPUs (Dragonfly C1000), custom silicon, and connectivity solutions. While these initiatives initially boosted its stock, concerns over AI capital expenditure sustainability have since erased those gains. The company targets $5B in data center revenue for FY2027 and $15B for FY2029. The report concludes that with the traditional handset business still under pressure, the data center opportunity is currently viewed as a longer-term option, and a more conservative valuation based on core operations may be warranted until AI contributions materialize.

marsbit1h ago

Qualcomm: AI Hype Subsides, When Will Smartphones Emerge from the Gloom?

marsbit1h ago

From TPU to Self-Evolving Agents: How Jeff Dean Predicts the Next Step in AI

At the 2026 YC Startup School, Jeff Dean outlined his vision for AI's next phase, shifting focus from simply scaling models to building intelligent, autonomous systems. He believes AI's progress is no longer just about creating smarter models, but about integrating them into systems capable of long-term, iterative work, automated experimentation, and continuous learning. This evolution moves the competition from "who has the bigger model" to "who can best organize intelligence." Dean suggests AI capabilities are now comparable to a junior engineer, enabling the automation of complex workflows. However, the true challenge and opportunity lie in managing these AI "workers" at scale. He emphasizes the importance of **context engineering**—structuring tools, memory, and feedback loops—over raw model power. For startups, this means building deep expertise in niche domains where general models currently fail (near 0-1% success rates), leveraging proprietary data, specialized tools, and domain-specific evaluators. A recurring theme is re-examining fundamental constraints. Dean's past work, like moving Google's search index to memory or creating the TPU, stemmed from questioning outdated assumptions about hardware and cost. He sees similar inflection points today, particularly in **specialized inference hardware** to drastically reduce latency and energy consumption for real-time Agent operation. Notably, he points out that in modern AI systems, the dominant cost is often not computation but **data movement**. Reliable, long-running Agents require robust system design, borrowing concepts from distributed computing like checkpointing, state management, and parallel exploration to handle failures and maintain progress over days or weeks. As AI automates execution, the scarcest human skills will shift to **defining clear specifications**, **judging what problems are worth solving** (taste), and designing effective feedback loops. Ultimately, Dean's framework prioritizes understanding the problem deeply, identifying the true bottlenecks, and systematically building closed-loop systems where AI can not only perform tasks but also improve AI itself.

marsbit1h ago

From TPU to Self-Evolving Agents: How Jeff Dean Predicts the Next Step in AI

marsbit1h ago

Trading

Spot
活动图片