GM. Welcome to Degen's Newsletter, your weekly shot of legal insights and regulatory updates in crypto and emerging tech. Our expert lawyers serve up curated analysis, breaking news, and expert commentary.

In this week’s edition:

1. Updates from around the world

2. Essential Insights

3. Regulatory Watch

4. Weekly Breakdown - This week we explain the term ‘Automated Market Makers’.

5. Article in the spotlight ‘Crypto-Friendly Countries to start a Web 3 Business in 2023’

6. Crypto Chronicles.

Updates from around the world

• Belarus plans to ban P2P crypto transactions citing cybercrime concerns

• Atomic Wallet faces class action lawsuit over $100 Million Hack

• South Korea passes Virtual Asset Protection Act to crack down on market crimes

• Thailand’s SEC implements investor protection rules for cryptocurrency service providers

Regulatory Watch

UK Financial Services and Markets Act, 2023

We know there's been a lot of noise about how UK will be affected by the new UK Financial Services and Markets Act, 2023 and the implication on both businesses and customers, hence, we decided to break down the most important points:

1. The law officially recognizes crypto as a regulated financial activity.

2. To market tokens to UK residents, registration with the FCA is mandatory (or) have someone who is authorized by or registered with the FCA approve the promotion.

3. Non-fungible tokens are not covered by the rules.

4. Promotions that induce someone to purchase tokens are prohibited.

5. Direct offer financial promotions require a 24-hour waiting period before sending the promotion, during which time recipient's name must be collected and KYC process must be initiated.

6. Reporting must be filed with the FCA covering metrics like how many users registered their interest, how many didn't go on to receive the promotion, and how many completed the whole journey.

Essential Insights

Safeguarding Privacy in the Era of AI: Part I

The US House of Representatives limited the use of OpenAI’s ChatGPT to the paid Plus version. Chat GPT was also banned in numerous countries like Italy ( temporarily ), China, Iran, companies like Apple and Samsung, and even platforms like StackOverflow! These actions stem from concerns over privacy and in same cases, information control.

In an era where personal data is digitized, easily accessible, and susceptible to abuse, data protection becomes crucial, especially in the context of emerging technologies like AI. Moreover, the privacy concerns regarding the use of Chat GPT, evidenced by incidence of breach such as the leak of over 100,000 ChatGPT credentials on the dark web, has shed light on the need for privacy preserving practices in the field of AI. In this context, there are several steps that may be undertaken by bot developers and platforms, users, and even regulators. We are going to break down the various concerns and sets forth measures which can be adopted to enable data protection and minimization of breach. The series shall be divided into the following sections:

1. General Safe Practices for Users and Developers

2. Technical Safeguards through Privacy Preserving Techniques to be Adopted by Developers

3. Regulatory Safeguards

4. Overview of Current AI Models and Ongoing Projects that Provide Safe Alternatives.

General Safe Practices for Users and Developers

As a developer or platform following are the steps you can undertake:

Privacy Preserving Techniques: While building bots, developers must use privacy preserving techniques. There are several techniques which can be used, such as differential privacy, homomorphic encryption, federated learning, etc.

Transparent data collection and use: Developers and platforms must be transparent about how they collect, use and share the data of their users. Consequently, they must provide clear information on these practices, users’ privacy rights and how they can control use of their data. Platforms must also have safeguards that give users rights over how the data may be used. For eg., Google facilitates users who use Bard to de-link their personal account.

Obtain user consent: Platforms must obtain user consent before collecting, using or sharing their data.

Securing data: Developers and platforms must take necessary steps to secure user data. Recently, there was a leak of over 100,000 ChatGPT credentials on the dark web. To prevent such breaches, there must be sufficient measures in place. Regulations like GDPR and PDPA, provide clear guidelines on how user data must be protected. Their application on data used in AI models will be dealt with in another blogpost.

Data breach policy: Platforms must have clear policies in place to handle data breaches and privacy incidents. For instance, regulations like GDPR mandates users to be notified of the same within a specific time frame. Unfortunately, not all jurisdictions are protected by adequate regulations. But, as a good practice you could adopt adequate policies to protect your users, even if regulations don't mandate it.

Even as a user, that are several steps you can take to protect your data, such as:

• Analyse the policies of the platform you are using, including the rights you have to control the use of your data. This will help you take necessary steps to protect your data.

• Limit sharing of personal data, that can help identify you or others.

• Keep your bot’s software up to date, to protect your data from security vulnerabilities.

• You can reduce the risk to a great extent if you disable chat history and training.

More on this in our next edition. Keep an eye out on the blog section of our page for more such in-depth insights.

Weekly Explainer

Automated Market Makers

In traditional finance, markets rely on order books where buyers and sellers set prices to govern market dynamics. In a similar manner in the world of crypto, ‘Automated Market Makers’ (AMM) help users exchange crypto by connecting one user to another directly without the need for an intermediary. AMM’s are used by Decentralized Exchanges to determine asset prices based on the ratio of assets in a liquidity pool. These pools are filled with tokens, and users can trade against the pool rather than with specific individuals. AMM’s have revolutionized liquidity provision in DeFi.

For example:

An automated market maker (AMM) is like a vending machine for trading. Instead of finding a specific person to trade with, you put your item in the machine, and it gives you another item with a fair exchange rate based on a set formula. The AMM helps people trade a candy for another candy without needing to find a specific person to trade with.

To learn more about such concepts, visit our resource page linked below:

Article in Spotlight

Crypto-Friendly Countries to start a Web 3 Business in 2023

With the MiCA law in the EU kicking in and the UK recognizing Crypto as Digital Assets as well as new rules in Dubai. Its time to revisit the most friendly crypto countries to set up Web 3 businesses as of July 2023, this list is not in any particular order and each country has its own relative merits.

1. Switzerland

The beautiful tiny country is well known for its banking & financial infrastructure. The Swiss Financial Market Supervisory Authority (FINMA) is responsible for the overall crypto-related activities carried out in Switzerland. It grants crypto licenses and monitors compliance with the regulations.

The Federal Act on the Adaptation of Federal Law to Developments in Distributed Ledger Technology stipulates rules for crypto licensing, trading, anti-money laundering, financial market infrastructure for trading cryptocurrencies and bankruptcy.

Switzerland classifies cryptocurrency as a private wealth asset – meaning that private investors are exempted from paying capital gains taxes on crypto & no income tax or wealth tax on cryptocurrency. For businesses, however, the Switzerland crypto tax laws are pretty much the same as the laws on legal tender or fiat currency.

It is also a great jurisdiction for businesses looking to start a DAO. Even though it does not have specific DAO laws, there are a couple of options including Swiss Foundations and Swiss Associations. Both have their own unique advantages, depending on the type of DAO and functionality required.

2. Singapore

The tiny city-state is one of the friendliest crypto jurisdictions in the world. It has a savvy financial services - the Monetary Authority of Singapore (MAS). The Payment Services Act (PSA) introduced in 2020 acts as the overarching regulation.

Digital currencies are referred to as digital payments tokens (DPTs), with Bitcoin (BTC) and Ether (ETH) recognized as cryptocurrencies by the MAS.

A Web 3 company operating out of Singapore must obtain a license under the PSA. Taxation in Singapore depends on the type of activity. Where trading in cryptocurrency is carried out in the ordinary course of business, the profit derived would be subject to income tax. Where it is purchased for long-term investment, capital gains would not be subject to tax. Singapore does however apply 8% goods and services tax on purchase, sale, or conversion of crypto assets.

3. Japan

One of the first mass adopters of crypto, japan still remains a desirable destination for those looking for clear regulations in running a Web 3 business. In 2017 it passed a law recognizing Bitcoin as a legal method of payment, making it one of the first countries to do so.

Recently, Japan's National Tax Agency has introduced new tax rules that exempt token issuers from paying corporate taxes on unrealized crypto gains.

In Japan, the crypto industry also has a certain amount of say in how it's governed. The Japanese Virtual Currency Exchange Association (JVCEA) is made up of more than twenty organizations that can pass and enforce regulations and standards for cryptocurrency exchanges in Japan. This model is unique and should be adopted by more countries.

To know more about other crypto-friendly countries and read the rest of this article, visit our website.

Tech Chronicles- Stories from the wild world of Tech

In this weeks edition of Tech Chronicles.

One of the most recognisable comedians in the world Sarah Silverman took on Open AI and Meta in a lawsuit before the Federal Court of San Fransico. The comedian and two authors have filed copyright infringement lawsuits stating that the companies used their copyrighted material to train artificial intelligence language models.

In an amusing twist that is sure to get ‘The Zuck’ riled up, in the lawsuit against Meta, the artists argued that they discovered their material was being used due to leaked information about Meta’s artificial intelligence language model LLaMA.

The first authors however are the first to sue OpenAI over copyright infringement, they currently face a host of legal challenges over possible violations of US federal and state privacy laws by scraping data to train the large language models behind ChatGPT and DALL-E.

Thank you for reading!

If you would like to receive regular updates, news, articles and emerging stories on law, cryptocurrency and technology, please click the subscribe button below!