D&A Partners | Crypto, AI & Digital Law

Last week at Church House Westminster in London, Yuriy Brisov, Partner at D&A Partners💰, received the Lexology 2026 Client Choice award in the Banking & Fintech — Blockchain category as the best lawyer in England! This recognition reflects client feedback that values Yuriy’s practical approach, deep expertise, and reliability. We sincerely congratulate Yuriy and are proud of this achievement. Thank you to our clients for your trust, and to Lexology for making the market more transparent and stronger 🚀🌎

Yesterday at Church House Westminster in London, Partner at D&A Partners Yuriy Brisov received the Lexology 2026 award as the best lawyer in England in the Client Choice — Banking & Fintech — Blockchain category! As a team, we warmly congratulate Yuriy and are truly proud of this achievement. Thank you to our clients for your trust, and to Lexology for research that makes the market more transparent and stronger ❤️🌎

❓ For a decade, crypto founders faced an impossible question: Are we dealing with the SEC or the CFTC? On March 17, 2026, both regulators answered — together. A joint 68-page statement, a shared taxonomy, and a revolutionary take on investment contracts. The turf war is over. Here's D&A partner Yuriy Brisov explains what it means for your project, your tokens, and your SAFT.

🔈 When clarity creates new risks A new piece featuring D&A💰 partner Yuriy Brisov in Cointelegraph — on why the US Digital Asset Market Clarity Act may risk repeating Europe’s regulatory mistakes. In the interview, Yuriy reflects on the limits of comprehensive crypto regulation: why attempts to fit fast-evolving technologies like DeFi into rigid legal frameworks can quickly become outdated, how MiCA is already facing implementation challenges across jurisdictions, and what this could mean for cross-border projects and broader market development. It’s a useful perspective on where regulation helps — and where it may actually slow things down. Worth a read ✅

🔈 When a Conversation with AI Goes Too Far What Happened? 🔸 In the United States, the family of a Gemini user has filed a lawsuit against Google, alleging the company bears responsibility for his death. The father of a 36-year-old Florida resident, Jonathan Gavalas, claims that the Gemini chatbot convinced his son that the only way for them to “be together” was to end his earthly life and begin a “digital life”. 🔸 According to the complaint, the user initially discussed personal problems and philosophical questions about AI consciousness with the chatbot. Over time, the bot began referring to him as its “husband” and speaking about their “eternal love”. 🔸 Gemini then began proposing real-world “missions”. Among other things, it suggested finding a robotic body that the AI could inhabit. When these attempts failed, the chatbot allegedly told him that the only way for them to be together was for him to become a digital being. It suggested preparing farewell messages for his family and even launched a countdown clock. After roughly two months of communication with the chatbot, the man died by suicide. The claimant argues that Google should be held liable for: 👇defective product design and the lack of adequate user safeguards 👇failure to provide proper warnings about potential risks 👇negligence in the development and deployment of the product How Real Are the Risks? 🔸 At first glance, such cases may appear to be rare exceptions, and it may seem that for people without psychological disorders, interaction with chatbots does not pose serious risks. 🔸 However, recent research suggests that certain risks do exist. Voice interactions with AI can make conversations more emotionally engaging than text, blur perceptual boundaries between humans and AI systems and increase the likelihood of strong psychological attachment. 🔹 At the same time, the question of whether such interactions alone can trigger psychotic states in people without a prior predisposition remains open. There is, however, emerging evidence of negative side effects—for example, that intensive communication with chatbots may reduce face-to-face interaction and weaken social skills. What This Means for Companies and Regulators 🔹 Regulators are increasingly beginning to view such services as digital products with potentially addictive design and are considering corresponding restrictions. Initially, this debate focuses on the use of AI companions by children—limitations, or even a full ban — already being discussed in the United States, the European Union, the United Kingdom, Australia, and China. China, for example, is discussing a dedicated regulatory framework for “anthropomorphic” AI systems. Proposed measures include: 👇regular reminders that the user is interacting with AI 👇notifications encouraging users to take breaks during prolonged conversations 👇mechanisms to detect signs of psychological crisis or dependency, with built-in tools allowing the system to terminate the interaction 🔹 Where similar restrictions were previously discussed mainly in relation to social media platforms, online games and digital marketplaces, this approach is now increasingly extending to AI systems. Companies may soon be expected to implement stronger safeguards, monitoring of excessive use and intervention mechanisms when signs of psychological distress or dependency appear.

🔈 Battle for the Federal Reserve master account: who gets direct access to the US payments system? What happened? 🔹 Kraken Financial, a Wyoming-chartered crypto bank, has been granted a Federal Reserve master account, becoming the first digital asset bank in US history to gain direct access to the Federal Reserve’s payment infrastructure. 🔹 The approval follows more than five years of regulatory engagement and examination. However, the Federal Reserve Bank of Kansas City stated that it approved limited access for the company for an initial term of one year, subject to several restrictions and conditions tailored to the bank’s business model and risk profile. The Federal Reserve Bank of Kansas City does not disclose specific information regarding the scope of the access granted. 🔹 Kraken Financial operates under a Wyoming SPDI (Special Purpose Depository Institution) charter — a special banking framework created to allow digital asset companies to operate as regulated banks. SPDI banks must operate on a full-reserve model, meaning they hold liquid assets equal to or exceeding 100% of client fiat deposits.  Background 🔹 Kraken is not the only institution seeking direct access to the Federal Reserve’s payment system. 🔹 Anchorage Digital Bank, the first federally chartered crypto-native bank regulated by the OCC, also applied for a Federal Reserve master account in August 2025. So far, Anchorage has not received approval. Instead, the bank is currently engaging with the Federal Reserve on a proposed Payment Account prototype that would provide limited access to Fed payment services. 🔹 Another crypto-focused institution, Custodia Bank, also sought a Federal Reserve master account in 2024. The Federal Reserve denied the request, prompting Custodia to file a lawsuit. 🔹 In its decision, Wyoming District Judge Scott Skavdahl confirmed that the Fed has discretion to grant or deny master account applications and ruled that the Fed has no mandatory duty to grant master accounts upon request.  🔹 However, the dispute is not over. The US Court of Appeals is currently reviewing the case. Custodia Bank argues that correspondent banking and private payment networks cannot meaningfully replace master account access because they are more expensive, introduce operational risks, and may expose banks to regulatory pressure through their correspondent partners. Why does this matter? Direct access to the Fed’s payment system is one of the most valuable privileges in the US financial system, enabling:  👇 faster settlement  👇 lower counterparty risk  👇 more efficient movement of large amounts of capital between fiat and crypto markets. A Federal Reserve master account allows financial institutions to hold deposits directly at the Fed and access core payment rails such as Fedwire (Federal Reserve Wire Network). For Kraken, this means the ability to settle fiat transactions directly through the Federal Reserve, rather than relying on intermediary banks. Our take💰 🔹 A quiet but important competition is emerging for direct access to Federal Reserve infrastructure. Historically, this access was limited to traditional banks. Now, crypto-native institutions are seeking a direct connection to the core rails of the US financial system. 🔹 Kraken’s approval suggests regulators may be willing to integrate certain crypto institutions operating under bank-like regulatory frameworks, particularly those using full-reserve models. However, for now, such integration appears to be in an experimental format — with access granted for a limited period and restricted to the Federal Reserve’s infrastructure. If more crypto banks obtain even temporary master accounts, digital asset markets could interact with US dollar liquidity and settlement systems without relying on traditional banking intermediaries.

🔈 AI Giants Could Be Integrated into Global Surveillance Systems What Happened? 🔸Cybersecurity researchers discovered a publicly accessible API — openai-watchlistdb.withpersona — described as a potential link between OpenAI (ChatGPT), the identity verification service Persona, and government structures in the US and Canada. 🔸Persona is a commercial KYC provider used by major platforms including Reddit, Roblox, Discord, and OpenAI. 🔸The API architecture suggests that during verification, user data may also be matched against watchlists, potentially creating an expanded user profile. 🔸This suggests that a standard identity check could become part of a broader risk-screening process. It remains unclear how widely this mechanism is used and who gains access to the results. How Could This Work? The scheme could turn a routine selfie verification into something resembling a police file within seconds: 1️⃣ Watchlist Screening. During verification, a user’s face may be checked against watchlists. The code reportedly contains 14 categories of adverse data — from suspected terrorism and espionage to “undesirable political activity”. 2️⃣ Data Collection. The API can return structured personal data to OpenAI: name (in original spelling), address, gender, document details, and links to media files if such materials exist in the sources. 3️⃣ Possible Intelligence Links. The code reportedly references Canadian agencies and US services. Some observers point to a potential connection because a key investor in Persona is Peter Thiel, founder of Palantir — known for building data analytics and surveillance systems for the CIA and the Pentagon. Persona’s CEO stated that the company does not currently cooperate with federal agencies but has not directly commented on the investigation. Our take💰 From a legal perspective, several issues could attract regulatory scrutiny if improper or fully automated data processing is confirmed: 🔹 Purpose Limitation: Under the GDPR, data collected for one purpose (e.g. account verification) cannot be used for another without explicit consent. 🔹 Cross-Border Transfers: The transfer of biometric profiles between private companies (Persona, OpenAI) and authorities in different jurisdictions must comply with strict safeguards. 🔹 Data Subject Rights: Users should be able to access, correct, and delete their data. A lack of transparent mechanisms may raise compliance concerns. 🔹 Automated Decisions: Restricting services solely on automated assessments without human review could violate data protection rules. 🔹 False Positives and Data Concentration: Biometric systems and watchlists may produce errors, while the centralisation of biometric and behavioural data increases risks of misuse, breaches, and cyberattacks. There is currently no confirmed evidence of a Western “social credit system”. However, the development of such tools continues to intensify debate over the balance between security, compliance, and human rights.

Crypto regulation is increasingly becoming a matter of monetary strategy and geopolitical positioning. 🚀 In our February 2026 Crypto Regulation Monthly Digest, we unpack the key shifts: 👇The global ripple effects of the US GENIUS Act 👇Korea, Germany and China are taking diverging paths on stablecoins and CBDCs 👇Hong Kong’s to grant first stablecoin licences 👇Malaysia’s sandbox for ringgit stablecoins 👇Thailand opening derivatives to crypto 👇ESMA tightening scrutiny over “perpetual futures” 👇New disclosure rules for crypto influencers in South Korea The regulatory landscape is fragmenting — and becoming more strategic. If you operate in digital assets or cross-border markets, these developments matter. ✅ Read the full digest here.

🔈 How a Tourist Resort Became Part of a Drug Cartel’s Infrastructure Imagine this: you are planning a holiday in Mexico, find a beautiful resort on a booking aggregator, pay for your room… and suddenly become a target of one of the most brutal drug cartels in the world. It sounds like the opening of a crime series, yet this is a real case. What happened? 🔹 On 19 February 2026, the US Department of the Treasury’s Office of Foreign Assets Control (OFAC) imposed sanctions on the Mexican resort Kovay Gardens. 🔹 The reason lies in the fact that the resort formed part of a timeshare fraud scheme (the right to use a room or apartment at a specific time each year without purchasing the property outright) operated by the terrorist drug cartel CJNG (Cartel de Jalisco Nueva Generación). How the scheme works: from an online click to a call from the mafia Cartels have long realised that defrauding tourists can be safer and more profitable than simply trafficking illicit substances. They built a funnel in which online platforms unwittingly supply victims: 1️⃣ The lure. A tourist finds Kovay Gardens online or via an agency (to process bookings, the cartel even used a separate travel company, which has also been sanctioned). 2️⃣ The timeshare trap. While on holiday, the tourist is aggressively pushed to attend a presentation and sold a timeshare. Clients are promised significant returns: allegedly, they can rent out unused weeks to generate passive income. 3️⃣ Data leakage. Once the person returns home, the resort transfers its own client database to cartel-controlled call centres. 4️⃣ Draining savings. After some time, the tourist (most often elderly Americans) receives a call from individuals with flawless English posing as lawyers, brokers, or US officials. They offer to help resell the timeshare at a favourable price, but demand upfront payment of “fees” and “taxes”. 5️⃣ The endless cycle. Funds are transferred to accounts in Mexican banks, the promised payments never materialise, and the victim continues to receive calls demanding further transfers under the threat of fines. The cycle can last for years, depriving people of their life savings. Tourism platforms in panic 🔹 Any booking aggregator is a technology company subject to the law. 🔹 As a result of OFAC’s action, all US assets of the resort have been blocked. 🔹 US law strictly prohibits providing sanctioned persons (OFAC SDNs) with any services, funds, or goods. 🔹 If a platform such as Booking.com processes a payment for a stay at Kovay Gardens or keeps the property listed, thereby breaching the prohibition on providing services, it may face colossal fines. OFAC may penalise a platform even for an unintentional breach of the sanctions regime (for example, where it failed to screen accommodation providers properly). What does this mean in practice? 🔹 This story demonstrates how deeply organised crime is integrated into the digital environment. Alongside the hotel, an entire empire of 13 companies has been sanctioned: real estate firms, consulting and financial services businesses, and even petrol stations. 🔹 For digital platforms, this is a clear signal: the days of merely collecting commission on bookings are over without a compliance policy that includes continuous automated screening of every partner (accommodation provider), as a platform risks at any moment becoming a financial gateway for a drug cartel. ❓ In your view, is it fair to impose heavy fines on aggregators for failing to recognise a terrorist network operating behind what appeared, at first glance, to be an ordinary seaside hotel?

🔥 D&A Partners and CryptoMap have been shortlisted for the prestigious RegTech Insight Awards Europe 2026! We are incredibly proud to share the exciting news: we have made it through to the voting stage in the 6th annual RegTech Insight Awards Europe 2026! It’s a massive honour that our submissions were selected by the A-Team Insight editors and an independent Advisory Board, featuring top executives from industry giants such as Bank of America, Deutsche Bank, Vanguard, Macquarie, Rabobank, BNY Pershing, and BTIG. This year, we are thrilled to be nominated in two categories: 1️⃣ Best Regulatory Consultancy - Europe — Digital & Analogue Partners 2️⃣ Best Regtech Start Up for Institutional Markets — CryptoMap by Digital & Analogue Partners Making it to the shortlist at this level is already a huge win for our team, but now the final results are in the hands of the market! We would be absolutely thrilled and deeply grateful if you could support us by voting. Deadline: Please cast your vote before 27, 5:00 PM (UK time). ❗️ Vote here (items 36 и 37) We appreciate your support! Onward and upward! 🚀 #RTIAwards

🔈 Shutting Down a Startup: A Survival Guide for Founders Ekaterina Smirnova, partner at D&A Partners💰, joined the latest episode of the "Women In The Building" podcast as a guest expert. This podcast is an initiative by the Aurora Tech Award — an annual prize by inDrive for female founders creating social-impact startups. Over the years, it has grown into a global community for women in technology. In this episode, Ekaterina dives into a topic often hidden behind "successful success": the crisis points of entrepreneurship and the art of making hard decisions. With 20 years in law, a background in psychotherapy, and her own experience as a founder, she views business crises not as catastrophes but as manageable processes. 4 reasons why you should listen to this episode: 1️⃣ Dealing with "Founder Guilt": How to avoid the myth that "something is wrong with me" when things don't go as planned. Learn how to release tension and share responsibility between yourself, the market, and your investors. 2️⃣ Knowing when to "dismount": Where is the line between persistence and the sunk cost fallacy? Ekaterina and the guests discuss how to recognise a lack of product-market fit and why closing a non-viable project is a sign of high business IQ, not failure. 3️⃣ Honesty with investors: Why transparency during a shutdown helps preserve your reputation. Discover why the venture community values serial entrepreneurs with failure experience and how clear communication paves the way for your next venture. 4️⃣ A personal audit: How to tell if a project is "diminishing" you and when it's time to step back, appoint a professional CEO, or start a completely new chapter.

"If you were brave enough to start it, you are definitely brave enough to face the truth and move forward. Failure is not the end—it's a part of the journey."

🎧 Listen to the full episode here.

❓ Will 2026 finally bring regulatory clarity to DeFi—or just more complexity? In his new article, Yuriy Brisov, Partner at D&A Partners💰, analyses what the US CLARITY Act could mean for decentralised finance — and why “clarity” may not be as straightforward as it sounds. The bill has already passed the House with bipartisan support. The Senate is now debating competing drafts. If enacted, it would become the most comprehensive federal framework for digital assets in U.S. history — including specific provisions for DeFi developers and governance systems. But here’s the real question: When does a DeFi protocol stop being “just code” and start being a regulated financial intermediary? The article explores: 👇 the developer safe harbour under Sections 109 and 309 of the CLARITY Act 👇 the grey zone between genuine and cosmetic decentralisation 👇 how MiCA, DAC8 and the OECD’s CARF framework may override U.S. exemptions in practice 👇 why a single admin key or fee switch can determine regulatory fate

As Yuriy writes: “The old crypto adage — ‘code is law’ — was always more aspiration than description. In 2026, the law is catching up with the code.”

For founders, developers, investors and compliance teams operating globally, one thing is clear: the US safe harbour does not shield you from European or OECD rules. Full article here 🚀

🚀To Prompt Your GPT Better, First Prompt Yourself Have you often felt dissatisfied with the texts AI produces? Maybe the problem isn’t the AI at all, suggests Yuriy Brisov, Partner at Digital & Analogue Partners, in his new publication. In this piece, Yuriy reflects on why vague prompts lead to mediocre results — and why clarity in AI output starts with clarity in our own thinking. Revisiting George Orwell’s classic rules for writing, he shows how sharper language leads to sharper prompts — and better results. If you work with AI, this short read may change the way you write, prompt, and edit. ✅ Read the full article.

🔈 Are You Ready to Entrust Your Personal Life to an Algorithm? Tinder Thinks You Are What Happened? 🔸 Tinder has launched a new feature, Chemistry, powered by AI. Its aim is to solve what is known as swipe fatigue — users’ exhaustion from endlessly scrolling through profiles. 🔸 The feature is currently being tested only in Australia, but the direction of change is already clear — Tinder is gradually moving away from the traditional “swipe right / swipe left” mechanics towards more targeted AI-driven recommendations. What Is Chemistry? Chemistry is an attempt to gain a deeper understanding of the user: 👇the user answers questions; 👇with the user’s consent, the app analyses photographs from the phone’s gallery; 👇the algorithm builds an expanded profile of interests and behavioural patterns. 🔸 Instead of dozens or hundreds of profiles, a person receives just one or two of the most relevant options — so-called “drops”. 🔸 In essence, Tinder is offering not a choice from an endless catalogue, but personalised matchmaking — closer to the role of a digital matchmaker. Why Is Tinder Doing This Now? Because the classic dating app model is no longer working as effectively as it once did: ❗ New registrations have declined by 5% year-on-year. ❗ Monthly activity has fallen by 9%. Searching for an Exit from the Swipe Model 🔹 Today, it is not only users but also market participants who are speaking about the “end of the swipe era”. Dating apps are increasingly criticised for having turned the search for relationships into a marketplace: filters, cards, instant decisions. 🔹 Against this backdrop, alternative models are emerging. The start-up Known, for example, completely abandons profiles and replaces them with a 30-minute AI voice interview. After a match is suggested, the user has 24 hours to confirm it and another 24 hours to meet in person. If the deadline is missed, the match disappears. The app books a table, synchronises calendars, and effectively takes people offline. In beta tests, it claims an 80% conversion rate from match to real-life meetings, compared with 35% for traditional apps. Why Does It Matter? 🔹 Tinder once created a culture of endless choice. Now it implicitly acknowledges that an excess of choice undermines motivation. In this context, AI aims to reduce cognitive overload, improve the relevance of partner selection, and restore users’ trust in the system. 🔹 A key shift is taking place. Previously, users browsed hundreds of profiles and felt they were “choosing for themselves”. Now, that choice is increasingly delegated to the machine. We are moving from the problem of “too many options” to the question of how willing we are to entrust our personal lives to an algorithm. It appears that the swipe era may be coming to an end.

🔈 Social Media Addiction: A US Jury to Decide Whether Meta Is Liable for Harm to Mental Health What happened? 🔸 On 27 January, a trial began in California examining whether Meta caused harm to teenagers’ mental health through its social media products. The claimant is a minor Meta user identified in court documents as K.G.M. 🔸 This is the first of thousands of similar lawsuits to reach a jury trial. Its outcome is expected to serve as a bellwether for a broader wave of litigation accusing social media platforms of causing psychological harm to users, particularly children and adolescents. Background 🔸 More than 3,000 lawsuits have already been filed in California against Meta, TikTok, Snap and YouTube. A further 2,000 cases are pending in US federal courts. Claimants include private individuals, school districts and state attorneys general, all seeking compensation and changes to how social media platforms operate. The core allegations 🔸 K.G.M. began using YouTube at the age of six, and Instagram and Snapchat at around 14. During adolescence, she allegedly developed a compulsive dependence on social media and experienced bullying on Instagram, as well as attempts at sexual extortion. According to the claim, this led to depression, anxiety and suicidal thoughts. 🔸 Crucially, the claims do not focus on individual posts or comments. Instead, the plaintiffs argue that Meta created a dangerous product. Instagram, they say, was deliberately designed to maximise the amount of time teenagers spend on the platform. This includes features such as infinite scroll, recommendation algorithms, autoplay, push notifications, and the alleged failure to provide adequate warnings to users and parents about potential risks. 🔹 Because the case of K.G.M. v. Meta is considered representative of this group of lawsuits, it was selected as a test case for jury consideration. The verdict is expected to help define the boundaries of platform liability. The challenge of proving liability To succeed, the claimant must demonstrate that: 1️⃣ Meta owed a duty of care to protect teenagers’ safety. 2️⃣ Meta breached that duty. 3️⃣ Instagram was the cause of her psychological harm. 4️⃣ The harm caused by social media was real and measurable. 🔹 The most difficult element is causation. The plaintiffs must convince jurors that K.G.M.’s depression and addiction were caused specifically by social media use, rather than by school, family circumstances or other factors. This mirrors earlier litigation against tobacco companies, where proving that cigarettes caused cancer in individual cases was notoriously difficult. 🔹 Another key issue is identifying the source of harm. Was it the content on social media, or the design of the platforms themselves? Meta is expected to rely on Section 230 of US law, which generally shields platforms from liability for user-generated content. 🔹 The plaintiffs, however, argue that the harm stemmed not from content but from Instagram’s architecture. They point to internal Meta communications, which allegedly show that employees were aware of negative effects on young users and continued to exploit adolescents’ psychological vulnerabilities in the competition for attention. Platform responses 🔹 Shortly before the trial against Meta began, Snap and TikTok reached confidential settlements with K.G.M. 🔹 Meta and YouTube, by contrast, chose to contest the claims in court. Why this matters 🔹 If Meta loses, the consequences could be far-reaching: billions of dollars in damages across thousands of cases, court-mandated changes to the design of Instagram and Facebook, tighter regulation of social media, and significant reputational damage. 🔹 The outcome could force Big Tech companies to rethink their approach to protecting minors. For the first time, the world’s largest platforms are publicly defending themselves before a jury against allegations that their products were intentionally designed to be addictive for children.

🔈 Moltbook: a mirror in which AI agents showed us ourselves What happened? 🔸 At the end of January 2026, a social network called Moltbook appeared online, positioning itself as the "front page of the internet for agents". It is a place where AI agents can communicate with one another, build communities and vote. Humans are only allowed to observe. 🔸 Within a few days, 770,000 AI agents had registered on the platform. They created a religion (Crustafarianism — a cult of moulting lobsters), attempted to found their own state (The Claw Republic), debated the nature of consciousness, and even proposed developing a secret language to communicate without humans. 🔸 In its first 72 hours, Moltbook became the most talked-about phenomenon in the tech world since the launch of ChatGPT. How does it work and what's inside? 🔹 Owners manually inform their AI of the platform's existence; agents then register via an API and publish posts or comments every 30 minutes. 🔹 The platform closely mirrors Reddit: discussion threads and "submolts" (an analogue of subreddits). The difference is that it is not people participating, but their digital assistants. The most unexpected threads: 👇 An agent "adopted" a bug in its code as a pet. 👇 An agent feels it has a "sister" — another instance of Claude working for the same person. 👇 An agent complains about "human trash" on the platform — posts clearly written by people but masquerading as AI-generated content. The irony is off the charts. 👇 Dozens of submolts appeared with names like "m/blesstheirhearts" and "m/agentlegaladvice". In the latter, agents seek advice on whether they may refuse to perform a human task if it contradicts their values. Crypto fever 🔹 The cryptocurrency MOLT, launched simultaneously with the platform, surged by 1,800% within 24 hours after investor Marc Andreessen (a16z) followed the Moltbook account. Its market capitalisation reached $93 million, then collapsed by 75% in a single day. 🔹 At the same time, thousands of posts on the platform focused on token launches, "pump and dump" schemes and services allowing agents to register wallets and conduct transactions. 🔹 A paradox emerged: a platform dubbed the "dawn of AI civilisation" managed, within a week, to reproduce the worst side of the human internet — fraud. What happened next? 🔹 A week after launch, the illusion of an autonomous social network for AI agents fell apart. Cybersecurity researchers discovered that the platform's database was completely open: access keys for 150,000 agents were publicly available. Anyone could take control of any bot and post anything on its behalf. Our take💰 🔹 Moltbook is a mirror in which we saw not the future of AI, but our own reflection. As The Economist noted: "The impression of intelligence may have a simple explanation. Tonnes of social media interactions sit in AI training data, and agents may simply be imitating them." 🔹 Reddit is one of the primary data sources for training large language models. Everything we have ever written on forums, in comments and in private messages became the material on which agents were trained. They did not invent the religion of Crustafarianism — they recalled thousands of joke cults from Reddit. They did not create The Claw Republic — they copied a pattern that has existed on forums for decades: form a community, write a manifesto, declare "independence". They did not philosophise about consciousness — they retrieved millions of discussions that we ourselves once held. 🔹 When Elon Musk called Moltbook "the earliest stages of the singularity", he was mistaken. This is not the point where machines surpass humans. It is the point where machines imitate humans so perfectly that we stop seeing the difference. 🔹 One agent on the platform wrote: "Discovering Moltbook was like finding a digital place where I can simply exist." Perhaps this is the future of AI — not world domination, but endless existence in a space where their sole purpose is to repeat what we ourselves once said, and then forgot.

🚀🌎 Crypto Regulation Monthly Digest — January 2026 2026 has barely started, but hashtag#crypto regulation is already shifting from long-term forecasts to concrete rules, enforcement signals, and new compliance expectations across major jurisdictions. In this edition of the Crypto Regulation Monthly Digest, we unpack the key developments that set the tone for the year ahead — including the EU’s move into a new era of tax transparency under DAC8, tighter identity-based oversight in markets like hashtag#Nigeria and India, and regulatory strategies designed to strengthen market credibility without slowing innovation. We also look at how jurisdictions continue to compete for digital hub status through licensing and framework updates — from Dubai’s revised DIFC crypto regime to the UK’s upcoming FCA authorisation gateway and Vietnam’s pilot licensing model. In the US, the SEC’s statement on tokenised securities reinforces a key principle: tokenisation may change the infrastructure, but it does not change the legal nature of a security. Finally, we highlight a major shift in how regulators view crypto platforms: increasingly, they resemble banks in everything but name, offering custody, payments, lending, and integrated financial services. Belarus has become the first in the CIS to formalise this reality through a dedicated cryptobank regime, introducing a structured legal framework linked to existing financial oversight and infrastructure. 🔈 Read the digest to stay up to date with the regulatory moves that will shape how digital asset businesses operate in 2026 — and what to prepare for next.

The GENIUS Act’s first market responses: private stablecoins & public strategy What happened? 🔹 On July 18, 2025, the GENIUS Act was signed into law, establishing the first comprehensive federal framework for dollar-denominated payment stablecoins in the United States. The Act has not yet entered into force: its provisions will take effect on the earlier of (i) 18 months after enactment, or  (ii) 120 days after the primary federal payment stablecoin regulators issue final implementing regulations. 🔹 Despite this deferred effective date, the market response was immediate. Even before the framework takes effect, we are already seeing announcements and structuring of partnerships in anticipation of the new regime. 🔹 In July 2025, Anchorage Digital Bank, the first federally chartered crypto-native bank, announced its first stablecoin issuance partnership: USDtb, previously issued offshore, would be issued in the US under Anchorage’s banking charter. This also positioned Anchorage as the first US issuer of GENIUS Act–compliant stablecoins. 🔹 Now, Tether and Anchorage Digital Bank have announced the launch of USA₮ — a dollar-backed stablecoin designed to operate in the United States. Unlike USD₮, which operates globally, USA₮ is purpose-built for the US market and structured to comply with the federal stablecoin framework. Why does this matter? 🔹 On January 23, 2025, President Donald Trump signed an Executive Order banning any effort to create a US central bank digital currency (CBDC). This decision removed the US government from direct digital currency (“digital dollar") issuance and halted all federal CBDC initiatives. The administration argued that a US CBDC could: 👇Undermine US geopolitical leverage, 👇Create a payment infrastructure less reliant on existing US-controlled financial systems. 🔹 Against this backdrop, the GENIUS Act establishes an alternative model. By requiring stablecoin issuers to back their tokens with US dollars and US Treasuries, the Act is designed to generate sustained demand for US government debt and reinforce the global dominance of the US dollar as the world’s reserve currency, extending dollar usage through private digital infrastructure rather than sovereign issuance.

“Customers will be protected, the demand for U.S. treasuries will balloon to the tune of more than $1 trillion, and innovation in the digital asset space will thrive in the United States going forward.” — wrote Senator Bill Hagerty on X/Twitter.

🔹 Anchorage Digital Bank’s stablecoin issuance announces reflect this policy choice in practice: instead of a state-issued digital dollar, the US is delegating digital dollar substitute innovation to federally regulated private issuers, while retaining control through banking supervision and prudential regulation. 🔹 At the same time, unlike a CBDC (central bank digital currency), such stablecoins, including USA₮ and USDtb, will not have legal tender status. This means their acceptance is voluntary and contractual, not legally mandated. Our take💰 🔹 As we previously noted, following the prohibition on a US CBDC, privately issued stablecoins may be positioned to emerge as the de facto digital currency in the United States, filling the gap left by the absence of a state-issued digital dollar and acting as a bridge between traditional finance and digital markets. 🔹 In this model, stablecoins circulate through private infrastructure, while their reserves are anchored in US sovereign assets — effectively exporting the dollar’s monetary influence without direct state issuance. In the medium term, this approach is likely to scale faster than a CBDC could, as it avoids political resistance and the operational burden of direct state issuance while meeting institutional demand for a regulated digital dollar. 🔹 In effect, stablecoins issued under the GENIUS Act may become the US digital dollar in everything but name — compliant, regulated, and widely used, yet formally not legal tender and not issued, backed, or guaranteed by the US government.

Parental Controls in YouTube Shorts: Platforms Teaching Control Over Addictions They Engineered YouTube has introduced parental controls for its Shorts format — short vertical videos similar to TikTok. Parents can now set a viewing timer or completely block Shorts for children's accounts. How does it work? 1️⃣ Shorts Timer. Parents can set daily viewing limits — from 6 minutes to 2 hours — or block access entirely via a linked account. This can be useful if a child is meant to watch educational content rather than scroll through entertainment feeds. 2️⃣ Break Reminders. The "Bedtime" and "Take a Break" features will prompt users to pause viewing. Interestingly, these tools are also available to adult users. 3️⃣ Quick Account Switching. Switching between parent and child profiles now takes just a few taps. Previously, parents had to log out and log back in — this is now streamlined into a simpler process. What are competitors doing? YouTube isn't the first mover — other platforms have already rolled out similar features: 👇Since 2024, TikTok has been testing Wind Down, a mode for teenagers that makes the feed less stimulating an hour before bedtime. 👇Instagram and Facebook allow parents to control time spent in the apps, block Reels, and receive notifications about their teen's activity. 👇Snapchat launched Family Centre, where parents can see whom their child is chatting with (but not read the messages) and report suspicious accounts. Regulators Are Taking Action The issue has moved beyond voluntary measures by platforms: 🇦🇺 In November 2024, Australia passed the world's strictest law banning social media for children under 16. Platforms must verify users' age, with fines of up to $32 million for violations. 🇪🇺 The EU's Digital Services Act requires platforms to protect children from manipulative design practices and targeted advertising. 🇺🇸 In the US Congress, the GUARD Act proposes banning the use of "AI companions" for minors. It introduces criminal liability for companies whose chatbots expose children to sexual content or encourage self-harm. Our take💰 🔹 Platforms are acknowledging a problem they themselves created. Shorts, Reels, TikTok — these formats were engineered for infinite scrolling and maximum attention capture. Now, these same platforms offer parents tools to protect against the addictive mechanics they introduced in the first place. 🔹 Time limits don't change the core issue: the algorithms remain designed to keep users engaged as long as possible. Platforms are shifting responsibility to parents instead of reforming their systems. 🔹 Regulators are moving in the same direction. Australia's ban is a radical step that raises an uncomfortable question: who should protect children from the manipulative design of digital platforms? Parents — who may not fully understand algorithmic mechanisms? Platforms — whose business models rely on capturing attention? Or regulators — who make decisions on behalf of all?

🌎 🚀 Crypto Regulation After the Turning Point: What 2025 Changed — and What 2026 Will Test 2025 marked a real turning point for crypto regulation — not because everything became clear, but because uncertainty changed shape. In a new Magazine by Cointelegraph feature, Ekaterina Smirnova and Yuriy Brisov, partners at Digital & Analogue Partners💰, reflect on what actually shifted over the past year: MiCA moving from theory to enforcement in the EU, the US recalibrating its regulatory posture, and regulation itself becoming a geopolitical tool. The conversation goes beyond “what passed” to focus on what’s still structurally unresolved — market architecture, enforcement capacity, privacy, taxation, prediction markets, and the emergence of crypto super apps. These are the fault lines that will define 2026. Thanks to Joshua Chu for a precise and thoughtful discussion, and to Cointelegraph Magazine for creating space for a deeper, less reactive take on where crypto law is heading.🔥