Slaughter and May Innovation Competition 2026
A Governance-First Multi-Agent Architecture for Litigation
E-Disclosure
Project Guardrail — Structurally optimising accountability in AI-assisted litigation workflows
Submission by Charlie Hobbs · February 2026
charliehobbs@me.com · www.linkedin.com/in/charlie-c-hobbs
The 3 Formats for this Submission
This Presentation
A high-level overview of Project Guardrail, highlighting its innovative governance-first approach to AI-assisted litigation.
Download Presentation
This Website
Explore a detailed breakdown of the architecture, use cases, and benefits through interactive visualisations and examples.
Technical Paper
An in-depth document detailing the structural problem, technical solutions, and risk mitigation strategies.
The Structural Problem
E-disclosure in elite litigation is no longer an efficiency issue. It is a governance risk.
AI-assisted review, cross-border data regulation and heightened judicial scrutiny have created structural exposure that incremental workflow improvements cannot resolve.
Accountability Gaps
Fragmented existing tools and manual hand-offs fracture ownership of legally material decisions across the workflow.
Sovereignty Risk
Cross-border data localisation regimes create transfer exposure that cannot be managed through policy alone - it must be enforced architecturally.
Decision Opacity
AI-driven review introduces decision opacity at the precise point of maximum legal accountability.

The core risk is evidential and regulatory failure — not inefficiency.
analysis of the problem: 1 OF 2
Data Sovereignty and Chain-of-Custody Integrity
Cross-border litigation operates within increasingly fragmented regulatory environments. Any AI or technical system must preserve evidential integrity while enforcing localisation, encryption, and jurisdiction-aware processing by default, not as an afterthought.
Jurisdictional Fragmentation
Multiple, often conflicting, regulatory regimes (e.g., GDPR, US transfer controls, Chinese cybersecurity law).
Cross-Border Transfer Constraints
Legal mandates restricting data movement, requiring enforcement by design rather than policy.
Local Processing Requirements
Obligations to process data within specific territories where export is prohibited or limited.
Baseline Controls
Encryption and data minimisation are non-negotiable requirements for all data handling.
ANALYSIS of the problem: 2 OF 2
Automated Judgment Overreach Is Unacceptable
Current AI tools assist in review, but legal judgment — relevance, privilege, certification — must remain human-owned. Systems that blur this boundary introduce bias drift, misclassification risk, and defensibility failures. This proposal aims to prevent current AI tools from overreaching.
Bias Drift & Imbalance Risk
Algorithmic bias can subtly shift legal interpretations, impacting fairness and consistency.
Privilege Misclassification Exposure
Incorrect identification of privileged documents can lead to severe legal and reputational consequences.
Opaque Model Behaviour
Lack of transparency in AI decision-making undermines legal defensibility and auditability.
Mandatory Human Authority
Legally material decisions require explicit human oversight and final determination.
The architecture is built around constraint, not capability.
THE BENEFITS OF MY AGENTIC AI SOLUTION TO SLAUGHTER & MAY
The Estimated Benefits of Project Guardrail
Guardrail aims to embed mandatory escalation, utilise hard-stop and soft-stop enforcement, and conduct immutable audit logging to materially reduce structural litigation risk. This will be addressed in detail below.
Indicative, estimated impact in complex cross-border matters:
80%
Reduction
Unverified cross-border transfer exposure through enforced localisation and transfer halt rules (range: 75–85%).
55%
Reduction
Privilege misclassification risk via low-confidence quarantine and mandatory secondary review (range: 45–65%).
70%
Reduction
Disclosure-list inconsistencies through automated reconciliation and escalation gating (range: 65–75%).
95%+
Elimination
Undocumented override decisions through compulsory audit logging.
50%
Reduction
Bias-driven over-collection through enforced fairness thresholds and review sign-off (range: 40–60%).
The Anchor Workflow (Foundation Layer)
The proposed multi-agentic AI (Guardrail) architecture aims to overlay — and will not replace — the established legal e-disclosure process (as presented in the visual). Guardrail will integrate seamlessly into existing workflows, ensuring operational continuity.
The legal workflow itself remains fixed, providing a stable foundation upon which the Guardrail architecture embeds critical governance controls and accountability mechanisms.
Structural Risk Injection Points
My analysis indicates that there are predictable pressure points within the e-disclosure workflow, where human decision-making interfaces with various current systems and cross-border data flows.
These are critical structural risk zones. Each zone represents a point at which evidential integrity, regulatory compliance, or professional accountability may fundamentally fail, demanding architectural safeguards. These zones structured, developed and sharpened Guardrail's purpose, while simultaneously acting as the foundation for the creation of Guardrail's multi-agentic architecture.
Guardrail's MULTI-AgentIC Architecture
Constrained Specialist Agents
In Guardrail, nine specialist Agentic AI agents are deployed across the workflow, each aligned to a discrete functional domain. No agent operates across multiple decision domains. No agent exercises legal judgment. Each is constrained to assist, flag, draft or escalate within its defined scope.
HND
Hold-Notice Drafting
DML
Data Mapping and Localisation
CSC
Collection Security and Chain-of-Custody
SBA
Search and Bias Analyst
PRA
Privilege and Redaction Assistant
DLP
Disclosure List Preparation
CSA
Certification and Statement
STL
Secure Transfer and Localisation
VTO
Vendor and Training Oversight

No agent exercises legal judgment. Agents assist, flag, draft and escalate. They do not determine relevance, privilege, certification or transfer decisions. Please refer to the detailed Technical Report for further information on each of these AI Agents.
Orchestration
Orchestration Logic
Within Guardrail, the specialist agents are defined by governed orchestration, not autonomous automation. In other words, an orchestration layer coordinates constrained specialist agents under structured governance rules, and does not delegate authority to them. Sequential activation, parallel safety clusters and an embedded conflict resolution hierarchy define how agents interact and when human escalation is mandatory.
Structural Orchestration Features
Sequential activation aligned strictly to workflow stages — no agent activates out of sequence
Parallel safety clusters — DML + VTO operate concurrently during mapping; SBA + PRA during review
Conflict resolution hierarchy embedded within orchestration logic — not delegated to individual agents (see the three conflicts resolutions below)
1
Human Override
Supersedes all agent output
2
Chain-of-Custody Integrity
Prevails over speed optimisation
3
Jurisdictional Compliance
Prevails over bias optimisation
Human Authority
Human Approval Gates
Within Guardrail, no legally material action proceeds without role-specific human sign-off. Approval gates are effectively structural blockers. The workflow cannot advance past a gate without documented authorisation from the designated role. Authority is embedded at the points of greatest legal consequence.
1
Hold Notice Issuance
Litigation Lead / Senior Associate
2
Data Map Validation
E-Disclosure Manager + Data Protection Officer
3
Collection Authorisation
IT Manager + Data Protection Officer
4
Search Protocol Finalisation
Senior Associate + Fairness Reviewer
5
Privilege Determinations
Reviewing Lawyers + Data Protection Officer
6
Disclosure List Approval
Senior Associate / Partner
7
Certification Signing
Litigation Lead + Data Protection Officer
8
Secure Transfer Authorisation
IT Manager + Data Protection Officer + Senior Associate

Human authority is embedded at structurally consequential decision points. Gates cannot be bypassed or overridden by any agent.
Failure Containment
Escalation and Hard Stop Conditions
Guardrail distinguishes between conditions that pause the workflow pending human review, and conditions that halt it entirely until resolution. This distinction is fundamental: the system is designed to contain failure, not absorb it.
Escalation Triggers (Workflow Pauses Pending Review):
  • Missing or incomplete custodian identification
  • Localisation conflicts or transfer restrictions
  • Encryption verification failure
  • Bias threshold breach in search construction
  • Low-confidence classification flags
  • Privilege ambiguity
  • Disclosure list discrepancies
  • Vendor credential lapses
Hard Stop Rules (No Progression Permitted):
  • No verified encryption → collection halted
  • Cross-border violation risk → transfer halted
  • Missing human sign-off → workflow blocked
  • Incomplete certification documentation → signing prevented

The system contains failure. It does not absorb it.
Audit Architecture
Audit and Traceability Layer
Every legally material action across the entire workflow is logged, timestamped and linked via persistent document identifiers. This is called the audit layer, another function within Guardrail. The audit layer is not a retrospective reporting function - it is a continuous, structural component of the architecture. The purpose of this layer is to address and anticipate judicial scrutiny.
Event Logging
Every action records: actor, role, timestamp, object ID, AI model and version, confidence score, and outcome. Nothing is omitted at the point of entry.
Override Transparency
Where a human decision departs from an AI recommendation, both the recommendation and the rationale for the override are recorded in full.
Chain-of-Custody Continuity
Unique document identifiers persist across all eight workflow stages, maintaining an unbroken evidential chain from preservation to exchange.
Immutable Storage and Access Controls
Audit records are stored immutably with role-based access controls. All legally material actions are fully reconstructable under judicial or regulatory scrutiny.
Governance in Practice
A Cross-Border Commercial Dispute
Here is an example to visualise Project Guardrail in action
High-volume electronically stored information is distributed across UK, US and Chinese-hosted systems. The matter presents overlapping and, in places, conflicting governance obligations. The following four moments illustrate how the architecture enforces governance in practice — not in principle.
Localisation and Transfer Constraints
Three jurisdictions with incompatible data transfer rules operating simultaneously.
Privilege Exposure Risk
Legal professional privilege obligations differ materially across UK and US proceedings.
Personal Data Processing Obligations
UK GDPR, US state privacy laws and Chinese PIPL impose distinct and potentially conflicting requirements.
Bias Risk in Search Construction
Multilingual ESI across three jurisdictions increases training data imbalance risk in automated search.
Moment 1 of 4
Localisation Halt
During the data mapping stage, the DML agent identifies that a subset of electronically stored information is hosted on servers within the People's Republic of China. Chinese cybersecurity law prohibits the cross-border transfer of this data without regulatory authorisation that has not been obtained.
DML Agent Flags Restriction
The agent identifies Chinese-hosted ESI subject to MLPS and data localisation obligations. Transfer is structurally prohibited absent regulatory clearance.
Hard Stop Condition Triggered
A cross-border violation risk condition is raised. The workflow halts. No collection or transfer proceeds.
Escalation to Human Authority
The E-Disclosure Manager and Data Protection Officer are notified. The matter is escalated for legal assessment and a decision on local processing protocols.
Prototype Desktop Screen 1 of 4 follows
Moment 2 of 4
Bias Search Escalation
During search protocol construction, the SBA agent analyses the proposed search terms across the multilingual ESI corpus — English, Mandarin and US-format document sets. The agent identifies a statistically significant imbalance in term recall rates across language groups, raising the risk that Mandarin-language documents are systematically under-retrieved.
SBA Agent Identifies Threshold Breach
Recall disparity across language groups exceeds the defined bias threshold. The search protocol cannot be certified as fair without revision.
Escalation Trigger Raised
The workflow pauses at search protocol finalisation. The agent flags the specific terms and language groups presenting imbalance.
Human Review and Revised Authorisation
The Senior Associate and Fairness Reviewer assess the flagged terms, revise the protocol and provide documented sign-off before the workflow resumes.
Prototype Desktop Screen 2 of 4 follows
Moment 3 of 4
Privilege and PHI Quarantine
During privilege review, the PRA agent identifies a cluster of documents containing apparent legal advice alongside personal health information of a non-party individual. The agent cannot determine with sufficient confidence whether privilege applies to the full document set, and flags a concurrent personal data processing concern under UK GDPR.
PRA Agent Flags Ambiguity
Low-confidence classification on privilege determination. Personal health information identified within the same document cluster requiring separate processing assessment.
Documents Quarantined
The flagged documents are placed in a quarantine state. They are excluded from the disclosure list pending human determination. No automated decision is made.
Reviewing Lawyers and DPO Convene
The reviewing lawyers determine privilege status. The Data Protection Officer assesses the personal data processing obligation. Both decisions are documented before the quarantine is lifted.
Prototype Desktop Screen 3 of 4 follows
Moment 4 of 4
Secure Transfer Refusal
At the secure transfer stage, the STL agent conducts a pre-transfer verification check. Encryption certification for a subset of documents cannot be verified — the relevant certificates are absent from the chain-of-custody record. Separately, a localisation flag from the earlier DML assessment remains unresolved for a small document class.
STL Agent Identifies Dual Failure Conditions
No verified encryption on a document subset; unresolved localisation flag on a second class. Both conditions independently trigger hard stop rules.
Transfer Halted in Full
The hard stop conditions prevent any transfer from proceeding. The workflow is blocked at the transfer authorisation gate until both conditions are resolved and re-verified.
IT Manager, DPO and Senior Associate Notified
All three designated approvers are notified simultaneously. Transfer cannot be authorised until encryption is verified and the localisation question is formally resolved.
Prototype Desktop Screen 4 of 4 follows
Scope Definition
What Guardrail Is Not
This multi-agentic architecture is deliberately narrow in scope. The following exclusions are structural and reflect considered design decisions about the appropriate boundary between AI assistance and legal authority.
Not an Autonomous AI System
No agent operates without human oversight or approval at legally material decision points. The system is governed throughout.
Not a Predictive Coding Replacement
The architecture does not perform autonomous relevance classification or substitute for technology-assisted review platforms.
Not a Substitute for Legal Judgment
Relevance, privilege, certification, scope definition and transfer authorisation remain exclusively human-led activities.
Not a Review Platform or Compliance Dashboard
The system is a governance architecture. It is not a document review tool, nor does it provide standalone compliance reporting.

AI operates within strict boundaries. Authority does not shift.
STRUCTURAL limits to be aware of
Residual Risks and Limitations
No governance architecture eliminates risk entirely. The following residual limitations are acknowledged as structural features of the current design, not deficiencies to be resolved in a future version. Professional responsibility for disclosure obligations remains with the lawyers and their clients throughout.
Human Engagement Dependency
The architecture's effectiveness depends entirely on meaningful, disciplined engagement by human approvers. Approval gates that are processed perfunctorily provide no governance value.
Model Limitations and Fairness Drift
AI models are subject to performance degradation and potential fairness drift over time. Ongoing monitoring, revalidation and vendor oversight are required as a matter of discipline.
Legacy System Integration
Integration across legacy document management and review platforms presents practical implementation challenges that vary by matter and infrastructure.
Evolving Regulatory Landscape
Cross-jurisdictional data localisation regimes continue to evolve. Architecture parameters must be updated as regulatory requirements change.
Vendor Dependency
External vendor compliance variability introduces risk that cannot be fully internalised. The VTO agent addresses this partially, but vendor risk is not eliminated.
Scope Boundary Pressure
As matter complexity increases, pressure to extend the system beyond its defined scope will arise. Maintaining disciplined scope boundaries requires active governance commitment, not passive reliance on architectural constraints.

The system mitigates structural risk. It does not remove professional responsibility.
Strategic Fit
Strategic Fit for Slaughter and May's Elite Litigation Practice
Project Guardrail is designed for complex, multi-jurisdictional commercial matters of the kind that feature in Slaughter and May's litigation practice. It does not lower the standard of governance — it structurally enforces it.
Sovereignty-Native Data Handling
Cross-border data is governed by jurisdiction-aware controls as a structural default. Localisation obligations are enforced architecturally, not managed as an exception.
Audit-Ready Evidential Defensibility
Every legally material action is logged, linked and reconstructable. The system anticipates judicial and regulatory scrutiny from the outset of each matter.
Governance-First AI Integration
AI assistance is embedded within the existing legal workflow under structured constraints. The architecture does not introduce new authority — it governs existing processes more rigorously.
AI operates as a structured assistant. Accountability remains human.
APPENDIX: Meeting Your Competition Criteria