In case you haven't noticed the consultation paper
, staff notice
, and report on Quadriga
, regulators are now clamping down on Canadian cryptocurrency exchanges. The OSC and other regulatory bodies are still interested in industry feedback. They have not put forward any official regulation yet. Below are some ideas/insights and a proposed framework.
- This framework was developed through a study of 96 past exchange hack/scam/fraud events. It prevents all historic cases where funds were lost in cryptocurrency exchanges.
- Typical securities frameworks will cost Canadians millions of dollars (ie Sarbanes-Oxley estimated at $5m USD/yr per firm). Implementation costs of this proposal are significantly cheaper.
- Canadians can maintain a diverse set of exchanges, multiple viable business models are still fully supported, and innovation is encouraged while keeping Canadians safe.
Many of you have limited time to read the full proposal, so here are the highlights:
Effective standards to prevent both internal and external theft. Exchange operators are trained and certified, and have a legal responsibility to users.
Regular Transparent Audits
Provides visibility to Canadians that their funds are fully backed on the exchange, while protecting privacy and sensitive platform information.
Establishment of basic insurance standards/strategy, to expand over time. Removing risk to exchange users of any hot wallet theft.
Background and Justifications
Cold Storage Custody/Management
After reviewing close to 100 cases, all thefts tend to break down into more or less the same set of problems:
• Funds stored online or in a smart contract,
• Access controlled by one person or one system,
• 51% attacks (rare),
• Funds sent to the wrong address (also rare), or
• Some combination of the above.
For the first two cases, practical solutions exist and are widely implemented on exchanges already. Offline multi-signature solutions are already industry standard. No cases studied found an external theft or exit scam involving an offline multi-signature wallet implementation. Security can be further improved through minimum numbers of signatories, background checks, providing autonomy and legal protections to each signatory, establishing best practices, and a training/certification program.
The last two transaction risks occur more rarely, and have never resulted in a loss affecting the actual users of the exchange. In all cases to date where operators made the mistake, they've been fully covered by the exchange platforms.
• 51% attacks generally only occur on blockchains with less security. The most prominent cases have been Bitcoin Gold and Ethereum Classic. The simple solution is to enforce deposit limits and block delays such that a 51% attack is not cost-effective.
• The risk of transactions to incorrect addresses can be eliminated by a simple test transaction policy on large transactions. By sending a small amount of funds prior to any large withdrawals/transfers as a standard practice, the accuracy of the wallet address can be validated.
The proposal covers all loss cases and goes beyond, while avoiding significant additional costs, risks, and limitations which may be associated with other frameworks like SOC II.
On The Subject of Third Party Custodians
Many Canadian platforms are currently experimenting with third party custody. From the standpoint of the exchange operator, they can liberate themselves from some responsibility of custody, passing that off to someone else. For regulators, it puts crypto in similar categorization to oil, gold, and other commodities, with some common standards. Platform users would likely feel greater confidence if the custodian was a brand they recognized. If the custodian was knowledgeable and had a decent team that employed multi-sig, they could keep assets safe from internal theft. With the right protections in place, this could be a great solution for many exchanges, particularly those that lack the relevant experience or human resources for their own custody systems.
However, this system is vulnerable to anyone able to impersonate the exchange operators. You may have a situation where different employees who don't know each other that well are interacting between different companies (both the custodian and all their customers which presumably isn't just one exchange). A case study of what can go wrong in this type of environment might be Bitpay, where the CEO was tricked out of 5000 bitcoins
over 3 separate payments by a series of emails sent legitimately from a breached computer of another company CEO. It's also still vulnerable to the platform being compromised, as in the really large $70M Bitfinex hack
, where the third party Bitgo held one key in a multi-sig wallet. The hacker simply authorized the withdrawal using the same credentials as Bitfinex (requesting Bitgo to sign multiple withdrawal transactions). This succeeded even with the use of multi-sig and two heavily security-focused companies, due to the lack of human oversight (basically, hot wallet). Of course, you can learn from these cases and improve the security, but so can hackers improve their deception and at the end of the day, both of these would have been stopped by the much simpler solution of a qualified team who knew each other and employed multi-sig with properly protected keys. It's pretty hard to beat a human being who knows the business and the typical customer behaviour (or even knows their customers personally) at spotting fraud, and the proposed multi-sig means any hacker has to get through the scrutiny of 3 (or more) separate people, all of whom would have proper training including historical case studies.
There are strong arguments both for and against using use of third party custodians. The proposal sets mandatory minimum custody standards would apply regardless if the cold wallet signatories are exchange operators, independent custodians, or a mix of both.
On The Subject Of Insurance
ShakePay has taken the first steps
into this new realm (congratulations). There is no question that crypto users could be better protected by the right insurance policies, and it certainly feels better to transact with insured platforms. The steps required to obtain insurance generally place attention in valuable security areas, and in this case included a review from CipherTrace. One of the key solutions in traditional finance comes from insurance from entities such as the CDIC.
However, historically, there wasn't found any actual insurance payout to any cryptocurrency exchange, and there are notable cases where insurance has not paid. With Bitpay, for example
, the insurance agent refused
because the issue happened to the third party CEO's computer instead of anything to do with Bitpay itself. With the Youbit exchange
in South Korea, their insurance claim was denied
, and the exchange ultimately ended up instead going bankrupt with all user's funds lost. To quote Matt Johnson in the original Lloyd's article
: “You can create an insurance policy that protects no one – you know there are so many caveats to the policy that it’s not super protective.”
ShakePay's insurance was only reported to cover
their cold storage, and “physical theft of the media where the private keys are held”. Physical theft has never, in the history of cryptocurrency exchange cases reviewed, been reported as the cause of loss. From the limited information of the article, ShakePay made it clear their funds are in the hands of a single US custodian, and at least part of their security strategy is to "decline to confirm the custodian’s name on the record". While this prevents scrutiny of the custodian, it's pretty silly to speculate that a reasonably competent hacking group couldn't determine who the custodian is. A far more common infiltration strategy historically would be social engineering, which has succeeded repeatedly. A hacker could trick their way into ShakePay's systems and request a fraudulent withdrawal, impersonate ShakePay and request the custodian to move funds, or socially engineer their way into the custodian to initiate the withdrawal of multiple accounts (a payout much larger than ShakePay) exploiting the standard procedures (for example, fraudulently initiating or override the wallet addresses of a real transfer). In each case, nothing was physically stolen and the loss is therefore not covered by insurance.
In order for any insurance to be effective, clear policies have to be established about what needs to be covered. Anything short of that gives Canadians false confidence that they are protected when they aren't in any meaningful way. At this time, the third party insurance market does not appear to provide adequate options or coverage, and effort is necessary to standardize custody standards, which is a likely first step in ultimately setting up an insurance framework.
A better solution compared to third party insurance providers might be for Canadian exchange operators to create their own collective insurance fund, or a specific federal organization similar to the CDIC. Such an organization would have a greater interest or obligation in paying out actual cases, and that would be it's purpose rather than maximizing it's own profit. This would be similar to the SAFU which Binance has launched, except it would cover multiple exchanges. There is little question whether the SAFU would pay out given a breach of Binance, and a similar argument could be made for a insurance fund managed by a collective of exchange operators or a government organization. While a third party insurance provider has the strong market incentive to provide the absolute minimum coverage and no market incentive to payout, an entity managed by exchange operators would have incentive to protect the reputation of exchange operators/the industry, and the government should have the interest of protecting Canadians.
On The Subject of Fractional Reserve
There is a long history of fractional reserve failures, from the first banks in ancient times, through the great depression (where hundreds of fractional reserve banks failed), right through to the 2008 banking collapse referenced in the first bitcoin block. The fractional reserve system allows banks to multiply the money supply far beyond the actual cash (or other assets) in existence, backed only by a system of debt obligations of others. Safely supporting a fractional reserve system is a topic of far greater complexity than can be addressed by a simple policy, and when it comes to cryptocurrency, there is presently no entity reasonably able to bail anyone out in the event of failure. Therefore, this framework is addressed around entities that aim to maintain 100% backing of funds.
There may be some firms that desire but have failed to maintain 100% backing. In this case, there are multiple solutions, including outside investment, merging with other exchanges, or enforcing a gradual restoration plan. All of these solutions are typically far better than shutting down the exchange, and there are multiple cases where they've been used successfully in the past.
Proof of Reserves/Transparency/Accountability
Canadians need to have visibility into the backing on an ongoing basis.
The best solution for crypto-assets is a Proof of Reserve. Such ideas go back all the way to 2013, before even Mt. Gox. However, no Canadian exchange has yet implemented such a system, and only a few international exchanges (CoinFloor in the UK being an example) have. Many firms like Kraken, BitBuy, and now ShakePay use the Proof of Reserve term to refer to lesser proofs which do not actually cryptographically prove the full backing of all user assets on the blockchain. In order for a Proof of Reserve to be effective, it must actually be a complete proof, and it needs to be understood by the public that is expected to use it. Many firms have expressed reservations about the level of transparency required in a complete Proof of Reserve (for example Kraken here
). While a complete Proof of Reserves should be encouraged, and there are some solutions in the works (ie TxQuick), this is unlikely to be suitable universally for all exchange operators and users.
Given the limitations, and that firms also manage fiat assets, a more traditional audit process makes more sense. Some Canadian exchanges (CoinSquare, CoinBerry) have already subjected themselves to annual audits. However, these results are not presently shared publicly, and there is no guarantee over the process including all user assets or the integrity and independence of the auditor. The auditor has been typically not known, and in some cases, the identity of the auditor is protected by a NDA. Only in one case (BitBuy) was an actual report generated and publicly shared
. There has been no attempt made to validate that user accounts provided during these audits have been complete or accurate. A fraudulent fractional exchange, or one which had suffered a breach they were unwilling to publicly accept (see CoinBene
), could easily maintain a second set of books for auditors or simply exclude key accounts to pass an individual audit.
The proposed solution would see a reporting standard which includes at a minimum - percentage of backing for each asset relative to account balances and the nature of how those assets are stored, with ownership proven by the auditor. The auditor would also publicly provide a "hash list", which they independently generate from the accounts provided by the exchange. Every exchange user can then check their information against this public "hash list". A hash is a one-way form of encryption, which fully protects the private information, yet allows anyone who knows that information already to validate that it was included. Less experienced users can take advantage of public tools to calculate the hash from their information (provided by the exchange), and thus have certainty that the auditor received their full balance information. Easy instructions can be provided.
Auditors should be impartial, their identities and process public, and they should be rotated so that the same auditor is never used twice in a row. Balancing the cost of auditing against the needs for regular updates, a 6 month cycle likely makes the most sense.
Hot Wallet Management
The best solution for hot wallets is not to use them. CoinBerry reportedly uses multi-sig
on all withdrawals, and Bitmex is an international example known for their structure devoid of hot wallets.
However, many platforms and customers desire fast withdrawal processes, and human validation has a cost of time and delay in this process.
A model of self-insurance or separate funds for hot wallets may be used in these cases. Under this model, a platform still has 100% of their client balance in cold storage and holds additional funds in hot wallets for quick withdrawal. Thus, the risk of those hot wallets is 100% on exchange operators and not affecting the exchange users. Since most platforms typically only have 1%-5% in hot wallets at any given time, it shouldn't be unreasonable to build/maintain these additional reserves over time using exchange fees or additional investment. Larger withdrawals would still be handled at regular intervals from the cold storage.
Hot wallet risks have historically posed a large risk and there is no established standard to guarantee secure hot wallets. When the government of South Korea dispatched security inspections to multiple exchanges, the results were still that 3 of them got hacked after the inspections
. If standards develop such that an organization in the market is willing to insure the hot wallets, this could provide an acceptable alternative. Another option may be for multiple exchange operators to pool funds aside for a hot wallet insurance fund. Comprehensive coverage standards must be established and maintained for all hot wallet balances to make sure Canadians are adequately protected.
Current Draft Proposal
(1) Proper multi-signature cold wallet storage.
(a) Each private key is the personal and legal responsibility of one person - the “signatory”. Signatories have special rights and responsibilities to protect user assets. Signatories are trained and certified through a course covering (1) past hacking and fraud cases, (2) proper and secure key generation, and (3) proper safekeeping of private keys. All private keys must be generated and stored 100% offline by the signatory. If even one private keys is ever breached or suspected to be breached, the wallet must be regenerated and all funds relocated to a new wallet.
(b) All signatories must be separate background-checked individuals free of past criminal conviction. Canadians should have a right to know who holds their funds. All signing of transactions must take place with all signatories on Canadian soil or on the soil of a country with a solid legal system which agrees to uphold and support these rules (from an established white-list of countries which expands over time).
(c) 3-5 independent signatures are required for any withdrawal. There must be 1-3 spare signatories, and a maximum of 7 total signatories. The following are all valid combinations: 3of4, 3of5, 3of6, 4of5, 4of6, 4of7, 5of6, or 5of7.
(d) A security audit should be conducted to validate the cold wallet is set up correctly and provide any additional pertinent information. The primary purpose is to ensure that all signatories are acting independently and using best practices for private key storage. A report summarizing all steps taken and who did the audit will be made public. Canadians must be able to validate the right measures are in place to protect their funds.
(e) There is a simple approval process if signatories wish to visit any country outside Canada, with a potential whitelist of exempt countries. At most 2 signatories can be outside of aligned jurisdiction at any given time. All exchanges would be required to keep a compliant cold wallet for Canadian funds and have a Canadian office if they wish to serve Canadian customers.
(2) Regular and transparent solvency audits.
(a) An audit must be conducted at founding, after 3 months of operation, and at least once every 6 months to compare customer balances against all stored cryptocurrency and fiat balances. The auditor must be known, independent, and never the same twice in a row.
(b) An audit report will be published featuring the steps conducted in a readable format. This should be made available to all Canadians on the exchange website and on a government website. The report must include what percentage of each customer asset is backed on the exchange, and how those funds are stored.
(c) The auditor will independently produce a hash of each customer's identifying information and balance as they perform the audit. This will be made publicly available on the exchange and government website, along with simplified instructions that each customer can use to verify that their balance was included in the audit process.
(d) The audit needs to include a proof of ownership for any cryptocurrency wallets included. A satoshi test (spending a small amount) or partially signed transaction both qualify.
(e) Any platform without 100% reserves should be assessed on a regular basis by a government or industry watchdog. This entity should work to prevent any further drop, support any private investor to come in, or facilitate a merger so that 100% backing can be obtained as soon as possible.
(3) Protections for hot wallets and transactions.
(a) A standardized list of approved coins and procedures will be established to constitute valid cold storage wallets. Where a multi-sig process is not natively available, efforts will be undertaken to establish a suitable and stable smart contract standard. This list will be expanded and improved over time. Coins and procedures not on the list are considered hot wallets.
(b) Hot wallets can be backed by additional funds in cold storage or an acceptable third-party insurance provider with a comprehensive coverage policy.
(c) Exchanges are required to cover the full balance of all user funds as denominated in the same currency, or double the balance as denominated in bitcoin or CAD using an established trading rate. If the balance is ever insufficient due to market movements, the firm must rectify this within 24 hours by moving assets to cold storage or increasing insurance coverage.
(d) Any large transactions (above a set threshold) from cold storage to any new wallet addresses (not previously transacted with) must be tested with a smaller transaction first. Deposits of cryptocurrency must be limited to prevent economic 51% attacks. Any issues are to be covered by the exchange.
(e) Exchange platforms must provide suitable authentication for users, including making available approved forms of two-factor authentication. SMS-based authentication is not to be supported. Withdrawals must be blocked for 48 hours in the event of any account password change. Disputes on the negligence of exchanges should be governed by case law.
Continued review of existing OSC feedback is still underway. More feedback and opinions on the framework and ideas as presented here are extremely valuable. The above is a draft and not finalized.
The process of further developing and bringing a suitable framework to protect Canadians will require the support of exchange operators, legal experts, and many others in the community. The costs of not doing such are tremendous. A large and convoluted framework, one based on flawed ideas or implementation, or one which fails to properly safeguard Canadians is not just extremely expensive and risky for all Canadians, severely limiting to the credibility and reputation of the industry, but an existential risk to many exchanges.
The responsibility falls to all of us to provide our insight and make our opinions heard on this critical matter. Please take the time to give your thoughts.
Digital money that’s instant, private, and free from bank fees. Download our official wallet app and start using Bitcoin today. Read news, start mining, and buy BTC or BCH. If assuming on common clients at BitRiver are utilizing comparatively extra state-of-the-art bitcoin mining machines, like Bitmain’s AntMiner S17s or equal models with an effectivity degree of about 50 watt per terahash second (W/T), BitRiver’s bitcoin mining farms might boast a complete hash charge of over 1,000 petahash per second (PH/s). The Bitcoin code is predetermined to halve this payout roughly every four years. It was reduced to 25 BTC in late-2012, and halved again to 12.5 BTC in the middle of 2016. You can keep track of the next Bitcoin halving on our Bitcoin clock page. Most recently, in May 2020, the third Bitcoin halving reduced the block reward to 6.25 BTC. What about transaction fees? The second source of revenue ... Investor, beware! TheStreet dishes the dirt on the top 7 bitcoin scams so far, and offers tips on how to avoid being duped. Binance is looking to consolidate more bitcoin mining hashrate to its pool in Russia and the Central Asia region.. The world’s largest crypto exchange is deploying a physical server node for its ... Cryptomining247 is a cloud mining provider that lets you access hash power over the cloud. Find out how it works today in our review. What Is Crypto Mining 24/7? Cryptomining247, found online at Cryptomining247.com, is a cryptocurrency mining service that offers cloud mining. With Cryptomining247, your mining hardware is already running: you pay the company a monthly rate, and they’ll ... BetterHash separates work information and pool payout information into different channels. The work information channel replaces Stratum as the communication channel between the Bitcoin network and the miner, and the pool payout information channel handles communication between the pool and the miner. This allows miners to choose their own block templates while the pool distributes payments ... Individuals, businesses, developers: learn from our simple Bitcoin guides. How Bitcoin works, what is Bitcoin, what is blockchain, how to buy Bitcoin, what is Bitcoin mining and more. Binance announced Monday the official launch of its mining pool service. In an announcement post, the company said that its initial offering would focus on bitcoin using Full Pay-Per-Share, or ... Hash Watch: bitcoin cash Miners Start Signaling Node Implementations. by Yuuma Nakamura. August 20, 2020. in Bitcoin. 0. 1.1k. SHARES . 3.9k. VIEWS. Share on Facebook Share on Twitter. There’s 88 days left till the bitcoin cash improve scheduled for November 15, 2020, and the group has been watching miners carefully. On August 18, the Bitcoin ABC workforce revealed a weblog put up that ...