From vending machines to global finance machinese1775ba6-2ad5-4aac... · cheat: “And then a...
Transcript of From vending machines to global finance machinese1775ba6-2ad5-4aac... · cheat: “And then a...
From vending machines to global finance machines
What are smart contracts?
Definition #1: a
machine
programmed with
rules we could
have defined in a
contract; instead
machine performs
or verifies
performance
What are smart contracts?
#2: two parties agree to run software
between them, or one party and the
blockchain
#3: long-lived blockchain process that
handles money & blockchain-titled
assets
“distributed application” or “fiduciary
process”
Tech often challenges legal categories
• “UberCab” vs. ride-sharing
• Bitcoin: currency or
commodity?
• Vaping: smoker or non-
smoker?
• “Are smart contracts
enforceable”? Is that even
a meaningful question?
Relationship between smart
contracts and traditional law
• A smart contract, like the repo man, is a security protocol that controls the burden of lawsuit
• If “possession is 9/10 of the law”, then a cross-border, blockchain-based financial smart contract may be 99%+ of the law
• Do smart contracts need to be legally enforceable? (Do courts needs to examine the guts of a vending machine to figure out what the parties intended?)
Smart vs. traditional contracts
Traditional contract: verbal text on paper text on computer – wet code interpreted by brain
Smart contract: rules / conditions analyzed & executed, as in vending machine -- dry code interpreted by machine
Some dry rules are more like rules from property, corporate, etc. deeds and charters, or from statutes or regulations, than from contracts: to make more distinctions with our metaphors we can think of “smart property”, “distributed autonomous organizations”, even “smart law”
Fiduciary (trusted) vs. fiduciary
process (smart contract)
If the public blockchain
has the money, it has
to treat it according to
its rules
Why smart contracts?
Secure a wide variety of desiderata
Costs of trust and security
Especially keen across borders: different
languages, laws, jurisdictions
Security = vulnerability minimization =
trust minimization
e.g. a financial transaction that minimizes
need to trust human institution(s) for its
integrity
Trusted strangers are security holes
• Tempting in
security protocol
design, but it’s a
cheat: “And then
a miracle occurs”
• Must be plugged
by highly evolved
institution or
blockchain
Minimize trust
assumptions
Decentralization per computer
science is much more automated
& secure than traditional security
Computer Science
Centralization is insecure
Insecurity drives up costs and gets you stuck
inside national silos
Biggest reason
traditional
finance is stuck
inside national
silos: it
depends on
governments
for its security.
"We appreciate the strong
partnership with law enforcement
in bringing the criminals to
justice” – J.P. Morgan spokesman
#JPMorgan Hack — Three Men
Charged in Biggest Bank Hack
in History (Identity Theft)
Public and global
Bitcoin Ethereum
Decentralization per computer
science is much more automated
& secure than traditional security
Computer Science
Diversity -> Independence -> Less Likelihood of
Collusion
Geopolitical insecurity (example: gold)
Geopolitically insecure finance
Russian holdings of Treasury bonds
Goldman Sachs collateral for DRs etc.
between U.S. and Russia: $100s of
millions
Visa & Mastercard for credit card
clearing between U.S. and Russia :
$100s of millions
Off on technicality: Argentine debt
default
Global Finance Machines
Security that:
• Doesn’t depend on politics, jurisdiction, or law
• Works seamlessly across borders: same rules in Albania and Zimbabwe, Estonia and Venezuela, China and the United States
Examples:
• Bitcoin: geopolitically secure currency• secure asset registry / title transfer
• Ethereum: geopolitically secure computer• secure financial instruments
Global Financial Machines
Examples and patterns
Dispute mediation
“Oracles”: input from the fallible world
Assurance contracts
Thought experiment: insurance
Dispute Mediation
2-of-3 multisig transaction between
2 parties
escrow/mediator
“Oracles”: input from the fallible
world On-chain programs can check for some
obvious kinds of fraud and error
Traditional source of fiduciary data: Birth and death certificates
Stock prices
Cryptography can protect integrity of transmission from source to blockchain
Fraud usually happens at the source (hard for oracles to protect against this) LIBOR scandal: manipulating the market itself
Forgery
Pools & assurance contracts
e.g. Kickstarter
Many parties send transactions to on-
blockchain pool
Blockchain code keeps track of total
When total > threshold funds
distributed to beneficiary
Expires without total > threshold
funds sent back to donors
Could it be the
lawyers that will
be replaced by
the robots?
Thought experiment: let’s try to replace the
insurance industry
Insured pays into blockchain pool
Insured makes claim
Claims adjuster(s) decide whether claim is valid & amount of damages damages released from pool to customer
Benefits: Public / transparent / auditable
Payments works seamlessly across borders (public blockchains only)
Don’t need permission of middleman: business is between the claims adjustor(s) & the insured
Problems with “smart insurance”
Problems:
Trustworthiness of & law governing claims
adjusters
Who sets rates & based on what actuarial
criteria (underwriting)
How are claims adjusters paid
Underwriting:
subjective (more
wet than dry)?
Claims
adjustment:
more wet
than dry
Interpretation & complexity
Dry code –> harsh results in presence of unanticipated exceptions
Anticipating edge cases: More conditions more complicated dry code
(software) & more reliance on fallible external data
Fewer conditions more role for human mediators more bias, whim, corruption, locality
Wet (traditionnal K) vs. dry (smart K) code
Law Software
Logic grounds
on
subjective
minds,
analogy
boolean logic,
bits
Security contempt/impr
isonment
replication +
cryptography
Predictability flexible rigid
Maturity highly evolved
/ many cases
larval / few
experiences
Wet (traditional K) vs. dry (smart K) code
Law Software
Area jurisdictional
silos
(on blockchain)
independence
from financial &
political
institutions &
seamless
operation across
borders
Costs lawsuits:
expensive
extremely low
Lessons from our thought
experiment
To take best advantage of blockchain,
we need to use best of blockchain tech
(public blockchains) with best of
traditional finance (claims adjustors,
stable investments, etc.)
Finance industry people who want to
operate with crippled blockchains (private
blockchains) are wrong
Purists think we can do everything with
public blockchains are wrong
Lessons from our thought
experiment
Smart contracts on blockchain provide
plumbing for financial aspects of service
which can seamlessly cross trust
boundaries (e.g. national borders)
Smart contracts don’t solve “wet”
problems (e.g. claims adjustment &
estimating risks -> setting fees)
Simplification?
Conclusion
Vending machine global finance
machine
Wet vs. dry code
Simplification?
Nick Szabo
Basic Copyright Notice & Disclaimer
©2016 This presentation is copyright protected. All rights reserved.
You may download or print out a hard copy for your private or internal use.
You are not permitted to create any modifications or derivatives of this
presentation without the prior written permission of the copyright owner.
This presentation is for information purposes only and contains non-
binding indications. Any opinions or views expressed are of the author and
do not necessarily represent those of Swiss Re. Swiss Re makes no
warranties or representations as to the accuracy, comprehensiveness,
timeliness or suitability of this presentation for a particular purpose. Anyone
shall at its own risk interpret and employ this presentation without relying
on it in isolation. In no event will Swiss Re be liable for any loss or
damages of any kind, including any direct, indirect or consequential
damages, arising out of or in connection with the use of this presentation.