ERAPreço de Caldera

Wall Street and global financial institutions are accelerating their entry into the digital asset market, blurring the lines between traditional finance (TradFi) and decentralized finance (DeFi). According to Bitwise CEO Hunter Horsley, the era of being "long on bitcoin and short on bank stocks" is over, as financial giants pivot towards embracing crypto. Horsley noted two asset managers with over $1 trillion in assets approved crypto products this summer despite a bear market, signaling widespread institutional belief in the industry's future. Sygnum's CIO, Fabian Döri, observes a structural shift where banks have moved from resisting digital assets to actively creating, supporting, and distributing them through custody, tokenization, and regulated trading. Anchorage Digital CEO Nathan McCauley adds that his firm's client base increasingly reflects the convergence of traditional and crypto finance, with large institutions preferring to partner with specialized crypto infrastructure providers rather than building in-house systems. Major firms like Swissquote, DBS Bank, BBVA, BNY Mellon, and affiliates of Credit Suisse, Morgan Stanley, and Charles Schwab have entered the crypto space in recent years.

The cryptocurrency industry is entering an era of regulated maturity, driven by frameworks like the CLARITY Act and the $GENIUS Act in the US. These provide clearer rules but also raise operational standards, requiring platforms to effectively demonstrate compliance. Simultaneously, fraud is evolving into sophisticated, AI-powered, lifecycle-based attacks, moving beyond simple onboarding scams. According to Daniel LaBarbera, VP of Sumsub for North America, companies must shift from one-time KYC checks to continuous, risk-based verification. This involves integrating identity, behavioral, device, and transaction data into a unified risk view. The 2026 State of Crypto Industry report highlights that while 55% of crypto firms faced fraud last year, effective strategies now combine AI-driven detection, continuous monitoring, and behavioral analytics. Key compliance challenges persist, particularly with the Travel Rule. Only 23% of companies are fully compliant, with high implementation costs, data security concerns, and regulatory fragmentation being major hurdles. Meanwhile, stablecoins are gaining traction, accounting for 36% of all crypto transactions in 2025, and evolving from trading tools into financial infrastructure for payments and settlements. The path forward lies in risk-based approaches that balance security, speed, and user experience. This includes adopting documentless verification, reusable KYC, and breaking down silos between compliance functions to create a holistic, lifecycle view of customer risk and trust.

On August 11, 2026, Anthropic announced that its new Claude models will embed invisible, machine-readable watermarks into generated text, allowing AI-generated content to be traced even after copying and redistribution, without affecting semantic quality or readability. This move follows global initiatives to watermark AI outputs, highlighted by the World Economic Forum and futurist Kevin Kelly. A core challenge is ensuring watermarking does not degrade model performance, leading to the key research goal of "provably lossless" generation. The theoretical foundation lies in provably secure steganography, which requires that watermarked data be indistinguishable from normal data in distribution. Early theoretical work was limited by the inability to precisely sample natural data distributions. The rise of generative AI models around 2018 provided a breakthrough, as these models learn and sample from explicit distributions. Pioneering work by a University of Science and Technology of China (USTC) team proposed frameworks for provably secure steganography using generative models, mapping encrypted messages to drive content generation. Subsequent research expanded the practicality of these techniques. Advancements include public-key steganography for asymmetric extraction, "box-less" extraction where the receiver doesn't need the original model, and "grey-box" scenarios for resource-constrained devices. A key innovation, SyncPool, resolved tokenization ambiguities in LLMs to ensure reliable message recovery. The principles of provably secure steganography directly enable "provably lossless" watermarks. If a watermark can be embedded without altering the model's sampling distribution, generation quality remains mathematically guaranteed. USTC's system, applied in platforms like Spark AI, offers plug-and-play watermarking for robust detection and model attribution. Internationally, research on "unbiased" or "distortion-free" watermarks for LLMs has accelerated, with work from Stanford, University of Maryland, and others. In the image domain, techniques like Gaussian Shading (CVPR 2024) embed watermarks by randomizing them into the latent space of diffusion models, proving no performance loss. Follow-up works enhanced robustness and addressed issues like parameter variation. The concept of "losslessness" has also been extended to model watermarking itself, using cryptographic constructs to ensure the watermark doesn't interfere with normal model use. The field has thus evolved from "empirical security" to "provable security," offering a theoretical guarantee against quality degradation from watermarking. The watermarking ecosystem involves an ongoing arms race between embedding and removal attempts, where the constraint of preserving content quality is a shared pursuit for both defenders and attackers, making the cost of removal a central part of the security博弈.

Following a period of heavy investment in 2020-2021, the cryptocurrency sector is reportedly seeing an accelerated shakeout of projects. Ryan Kirkly, CEO of Global Settlement Network (GSN), states that projects with high valuations but unsustainable revenue models are now leaving the market. Since early 2026, over 100 crypto and blockchain projects have ceased operations or gone bankrupt, a trend largely attributed to the earlier investment frenzy. While venture capital investment in Q1 2026 saw $4 billion across 355 deals, the total sum has halved from Q4 2025, primarily due to fewer mega-rounds. Kirkly believes the ongoing market cleansing will strengthen sectors like stablecoins, digital banking, institutional wallets, and payment infrastructure, while social tokens, memecoins, and some Web3 gaming startups may face greater pressure. He notes sustained corporate and government interest is shifting towards use cases like cost reduction, asset tokenization, and cross-border payments. Regarding Bitcoin, Kirkly describes the outlook as a "moderate bear market," identifying $61,200 as a critical support level. A break below this could trigger leveraged fund liquidations and potentially push prices toward $41,000.

The recent spate of incidents involving Coldcard, Trezor, and SafePal highlights a critical evolution in cryptocurrency wallet security, moving the focus beyond simple private key protection to a holistic, multi-layered attack surface. These events—spanning a random number generator flaw, supply chain data leaks, and plugin permission issues—underscore that vulnerabilities now exist across the entire wallet lifecycle: from secure element and code generation to logistics, user data, and daily interactions with dApps. This broadening threat landscape is accelerating with the advent of AI. Attackers are leveraging AI to automate and scale previously labor-intensive tasks like vulnerability discovery, sophisticated social engineering, and targeted phishing campaigns. This effectively lowers the cost of attacks, eroding the security margin once provided by the high effort required to find and exploit flaws. In response, defense strategies must also evolve by integrating AI. The future of wallet security lies not just in static rules and blacklists, but in proactive, AI-powered risk assessment. This includes pre-transaction simulation, behavioral analysis to detect anomalies (like sudden large approvals), and contextual awareness of dApps and counterparties. The goal is to transform wallets from passive signing tools into active guardians that can understand intent, predict outcomes, and clearly communicate risks to users—all while preserving user sovereignty and control through minimal permissions and human confirmation for critical actions. Ultimately, self-custody does not guarantee inherent safety; it returns absolute control to the user. Protecting that control requires a dynamic, evolving security posture where AI becomes a essential tool on both sides of an ongoing "spear and shield" arms race in the Web3 ecosystem.