Papers

Live Appraisal of Key Compromise via Duplicity Evident Infrastructure How to protect verifiers (and controllers) from imposters without blockchain and trusted third parties r

& RERI

somuets oo, [T bt

Definitions

Risk Assessment: Internal process whereby a Controller assesses its own infrastructure to determine likely vulnerabilities to exploit and actual exploits and compromise. Static Risk Assessment: Systemic Vulnerabilities Prior to Attack Dynamic Risk Assessment: Pre and post (mostly) evaluation and evolution of suspected attacks.

Largely forensic after the fact (lagging vs. leading indicator of compromise. “tracing the untraceable” Appraisal: External process whereby a Validator appraises some other Controller’s infrastructure to determine likely vulnerabilities to exploit and actual exploits and compromise. Static Appraisability: External Validator has visibility into the systemic vulnerabilities of some other Controller’s infrastructure Dynamic Appraisability: External Validator has visibility into the live state of exploit (compromise) of some other Controller's infrastructure

IETF-Remote ATestation Procedures (RATS) https:/datatracker.ietf.org/group/ratsjabout/ It prtocol nchanges i o hecas st v iy o g Party) rocio vidanco bt th o oot (and sytancompanerts [FECAS4S] o). e o 45502 h shearina f I . ot e alon IoCatues (RTS) oo whehor obing s can bl vl o conince i1 userininse f ol pors, calod Ao T Ol STovi - 23 3p L5l pocedurs GEHaloa by & o IV P cales Vot i i ks o e g S T procacres for e two siges v

• Edonce Appeisl Vi appis pocyan supply chainput, s as Endorsemens and References Vs o e Atesaion Resls rom Eidnc.

• tsaton Resuits Aprasl s Faying Partyappespocy 1 Atestaion Res'ssssoiated it an At Evidece tht ofinales o a sted Verfes The

et it Gt iy A

To mprov theconidonce i sl Component ustwortiness, roig party ey reui okl abnt

• sysiem component dnty.

" Composton of syt coponent, naing nsted corporiets,

pstigetiol

" aserioiciim orgnaton o provarance,

andachain o,

st component irty

et componentconasaion,

. araion i i st of st Vch e th oartoal i, o

• other fctors st couks s s oions, W o ststation macariss s s T Corputing Goup (1) Tt Pt o (TP TPM Softe Stack(15S),Fat ity Onine (FIDC) Allnce atstaton 2 AT Keytor aetaon Kt er .o itopaable a1 Tt races astaton sidence 10 ks detrintons sboct St componets amon e Tt o e MEACIS a1 TRt Ao ProcEursS (RATS) N datatracher o rouBaryaboss

KERI and Dynamic Appraisability Duplicity Evident infrastructure dynamically (in near real-time) ensures that evidence of both live and dead compromise of key state is avallable to any validator. Any validator can then perform a live appraisal of key state compromise before engaging in any trust task. Live appraisal can be staged (graduated) to match the level of risk with the degree of evidence supporting the appraisal prior to committing to the trust determination. Duplicity Evident Infrastructure removes the hard reliance on trusted third parties to perform live appraisals. This s the primary security innovation that KERI as a DKMI provides. It s the simplest known approach to solving the dynamic appraisability problem without relying on trusted third parties or blockchain. Duplicity Evident(KERI) vs. Duplicity Hiding (blockchain) vs. Duplicity Fostering(DNS/CA)

End Verifiability End-to-End Verifiability - Midde »- P dta [Mdm ( A’ Yo} - [ Middie - o data 7 el daa e » A’ — N End L 4 Middie > data ——+ Middle »——— e — End Mo’ * Middle »— 4« { dai2 . Middle + If the edges are secure, the security of the middle doesn’t matter. Ambient Verifiability: any-data, any-where, any-time by any-body Zero-Trust-Computing It’s much easier to protect one’s private keys than to protect everyone else’s internet infrastructure

Cryptographic Root-of-Trust: Self-Certifying Identifier (SCID) + Key Event Log + Infrastructure = Autonomic Identifier (AID) oot —, i + Oevetr | gD Foiorset - s Py Geesn - ek vt ¥ | P Fokorwt - s - ety + G - Ak S o ¥ 3 Key Event Log Raensood > Sy iy Geten " Py | A noepon erigaton)— i prooi iy e cspoon et Icopten Event otonEvrt RotaionEvent Incepion Satement sonEert Ineracin Event ooectn Dety RatonEvort nracon Event Dertaion | Publkeys | Confiuaton | Signatres P RotaionEven AID Inerasen Event e ; RotaionEvert FXq5Yqal.6148pFOfu7TURL0IRAD2_RxFPOALA3wYn148 — Name-spaced AID Ifeacton Event did:keri:EXg5YqaL6L48pL0fuTIUNLOJRAU2 RxFPORLA3WYn148/path/ Lo/ resourcenane=securefreally

Inconsistency and Duplicity inconsistency: lacking agreement, s two or more things i relation to each other oy SRS wen duplicity: acting in two different ways to difierent people concerning the same matter it | Internal vs. External Inconsistency oo et neracton Event Internally inconsistent log = not verifiable. Interacion Event Log verification from self-certifying root-of-trust protects Rolation Evert | against internal inconsistency. Rotaon Event. | Inraon Evont | Externally inconsistent log with a purported copy of log but ol Event FE ey both verifiable = duplicitous. Interaction Evert Duplicity detection protects against external inconsistency. KERI provides duplicity evident DKMI

KERI Ecosystem Components: Controller Application and Agents (" Controller | Application key pair generation key pair storage key event generation el key event signing d‘b key event validation (" Controller | . Controller Agent | . Application key pair generation key pair storage key event generation key event validation E key event signing A ; J ) Modular, decentralized, web-based infrastructure without shared governance.

KERI Ecosystem Components: Peer-to-Peer Direct Mode K i "~ Contoller | " Controller | oy vent signng foyevnt generaion ¥ @ fi e event Sgning oo o’$o ey pair sorage Controller Agert | Contoler Agert | L events—— ey par storage key event validation events key event validation 5 )

KERI Ecosystem Components: Witnesses and Watchers, Indirect Mode Key pair generation key event vaidation ey event vaiidation Key pair generation ey event generation key event receipting key event receipling Key event generation ey event signing ey event publicalion Key event publication ey event signing Controller | Witness Pool | Witness Pool Controller Application — Appiication @ QD eens D fi o'y0 — — 4o I events events I Conlroller Agent [ Watcher Pool Watcher Pool Controller Agent | r— M events 060 events 000 events key pair storage ey event vaidation ey event vaiidation ey pairstorage Key event vaidation ey event vaiidation Modular decentralized web based infrastructure without shared governance

KERI Ecosystem Components: Witnesses and Watchers, Indirect Mode [ — —— - [ — oy ot ginrton o vaniring oy mantrcsing o gmrston ey even s oyt eiton Loy o ptaion ey g Caroter f—— SirassPa Carelr Application Appiication [] anm an [a) o4 [ m—] — L7 $ v iy et o o &_\ o2 o2 ’ o o= S5 s G- B gt - 7 ool e senss ovens i [ - ot owts b s - | Vostertiork | Vet <‘ ‘o% VatberFos Q0. 8 . Femd | OO0Q [ an oy vt (e ) ey avent vlidation upily detackion ey avent validation ‘Ambient Verifiability

KERI Identifier KEL VDR Controls Verifiable Credential Registry TEL VDR D TR R —— T sy i o d T o Tarmackn Koy Bt i o] on | e o Lo [ |- rmain] o] s o oo |-+ Tomton ||+ 1| o0 | opwatin | 0. ] 2| o | st b Tt 2] sot | oman | smw [ T3] e | s ~ Toansscon |+ 3| seal | Opsrston e J‘ [(Soa | o 4] S | O Sm ‘Seal Tonsscion |+ 5| Sesl | Operan St | AAKERIKEL for a given identifier provides proof of authoritative key state at each event. The events are ordered. This ordering may be used to order transactions on some other VDR such as a Verifiable Credential Registry by attaching anchoring seals to KEL events. Seals include cryptographic digest (SAID) of external transaction data that binds the key-state of the anchoring event to the transaction event data anchored by the seal. The set of transaction events that determine the external registry state form a log called a Transaction Event Log (TEL). ‘The transactions likewise contain a reference seal back to the key event authorizing the transaction. This setup enables a KEL to control a TEL for any purpose. This includes what are commonly called “smart contracts”. ‘The TEL provides a cryptographic proof of registry state by reference to the corresponding controlling KEL. Any validator may therefore cryptographically verify the authoritative state of the registry.

Trust Basis of a Trust Domain | controller ‘ » snuce;)'—t'uh . ,'/ B AN verifies bindng 1900 g Y system \ seautty N [ vaiidator e : overlay ) . | key-pair |« binding— | identifier |

Administrative Trust Basis DNS/CA, OIDC IP | controller | generates N/ sourceofuh |

/ database

stong .

binding — h

/ assigns. manages validator key-pair | | identifier | administrator | root-of-trust in non-verifiable operational infrastructure with opaque governance

Algorithmic Trust Basis Shared distributed ledgers

_ controller |,

£ N registers source-oHtruth generates / \ ] shared ledger |+

\ / v e veifies ~ strong N strong A \

binding “e'l“es binding . \ . / registers \ manages _ validator |

| keypair — :‘:‘;‘""%7% identifier | algorithm | root-of-trust in verifiable operational infrastructure with shared governance

Autonomic Trust Basis Cryptographic proofs via verifiable data structures N manages _ 1 controller | ~ N prows source-of-tuth generales /‘ ‘\ | key eventlog |*~ Y 4 veriies strong strong \ binding cerives binding N & B / o \ B validator | Key-pair |« 5209 ——=| identiier root-of-trust in verifiable cryptographic proofs of infrastructure with no shared governance

Use Case Example: Automating the “Signing” of a Mortgage Bank “Bob” (Lender/Mortgager) wishes to load funds to Individual “Ida” (Borrower/Mortgagee). How does Bob utilize the duplicity evident infrastructure of Ida to make a live appraisal of compromise of Ida’s key state before transferring funds? Identity assurance Accountability/Liability incentives for Ida to be malicious vs. impersonated Ida digitally commits to the mortgage agreement by binding (sealing/anchoring) SAID (hash) of the agreementina TEL that is in turn bound (sealed/anchored) to Ida’s KEL, thereby binding the mortgage transaction event o Ida’s current key state. da presents to Bob a bound mortgage agreement and proof of binding. Bob performs live appraisal to determine evidence of duplicity. * Verifies TEL, KEL, and Mortgage Agreement bindings * Characterizes strength of bindings and the relative likelinood of compromise (multi-sig, witness, rotation vs interaction event, delegated vs non-delegated) *Checks its watchers of Ida’s Witnesses; is there any evidence of duplicity? *Waits for some time (some meaningful multiple of network propagation time) for a recovery rotation by Ida that invalidates bindings. This enables Ida Watcher's ongoing live appraisal to detect and recover from a compromise of her owninfrastructure.

Backup Slides

Setups Matter in Layered Security Overlays Authenticity Authenticity — e = - Sonsetn i rx/' | )\ 008 setp Pt . R\ \ g St BIX H p E dst 2 dst & WA : JTTTs 1 & its) 3 \_y & - % § o5 Layered Security Overlay Spiral Setups may be application dependent & hence overlay ordering dependent Privacy Each setup requires one OOBA factor to protect against MITM attack. Layered Se Overlay Spiral Setups may be costly, especially those with repeated OOBAs ayered Security Overlay Spira Interactive setups are not as scalable as non-interactive setups Trade-offs required

KEL Anchored Issuance-Revocation Registry with Separable VC TELs eyt R S . o et Ve G iy o do T T ot ion Jr— o s fints Sepe U Taracn B T Sew oo VoS 1 ks | s || Tamacin | (RN sea || | Tonwcion | (1 sl | R | fonee {2] oo [ st ] i | Soprl VC Traracton Evt s T on [ s [Tt} T e Ol VOSHe o o e e sl Transacion -1 sl Revoke | Revoked R — — Sopatio VO T T ek Comrs WCSuw o s [ e [ e | | s | e | e Each VC has a uniquely self-addressing identifier (SAID) Each VC has a uniquely identified issuer (AID) Each VC may have a uniquely identified issuee (AID). AllVC Schema are immutable

Identifier Delegation: Scaling & Protection = | e ‘M = e sl wmml e i i

Hard Problems & Solutions Moving Data Across Trust Domains. Trust Spanning Protocol (TSP)(SPAC) No Shared Secrets Authentic Chained Data Container (ACDC) No passwords Key Event Receipt Infrastructure (KERI) No shared encryption keys Composable Event Streaming Representation (CESR) No bearer tokens GLEIF vLEI No shared private keys Key Management (rotation) True Zero-Trust = Sign Everything Global Portability At-Scale it micire it s Pl oo s o 8 - O 8 - O 8 - O 8

Universal Secure Attribution Problem Establish authorship of data, documents, credentials, entitlements, .. = Verifiable secure attribution of any communication to its source = Authentic data provenance by anyone to anyone from anyone Solve data provenance to solve security l controlling key(s) ‘ —— secure ‘ identifier ‘«— authorship —r) E— attribution

Identity (-ifier) System Security Overlay Establish authenticity of IP packet’s message payload. o Authenticatable Message Verifiable Payload {identmer }- { identiy system mapping |+ - key-pair \ The overlay’s security is contingent on the mapping’s security.

Cate promises to provide a Duplicity Game How may Cate be duplicitous consistent pair-wise log. and not get caught? Local Consistency Guarantee g Cate Contoter Log / B A LogV1 LogV2 LogV2 a = AP 3 aidaor [l Log Log private (one-to-one) interactions