Access to Justice Is Real -- But Smaller Than the Hype
The honest version of the story is better than the exaggerated one. AI can expand access to justice. It cannot rescue the justice system from all the reasons it fails people.
The 7.50 GBP letter before action has become the symbol of AI's promise to democratize legal services. Garfield, the first SRA-authorized law firm whose offering is solely provided by AI, has produced something the sector rarely manages: evidence that a legal service can become radically cheaper without becoming useless.
That matters. It just does not mean what the loudest headlines say it means.
Garfield is proof that AI can make specific, standardised legal interventions economically viable for people who would otherwise get no help at all. It is not proof that AI has solved access to justice as a system.
The Garfield Model
In 2025, the Solicitors Regulation Authority authorized Garfield as the first AI-only law firm in the UK. SRA chief executive Paul Philip called it "a landmark moment for legal services in this country."
The pricing structure challenges traditional economics:
- Polite chaser letter: 2 GBP
- Letter before action: 7.50 GBP
- Full small claims representation: 100 GBP + VAT
Compare this to the Channel 4 Dispatches experiment, where traditional legal services quoted 1,080 GBP for the same work that Garfield completed for 100 GBP + VAT. A senior solicitor judging the blind comparison deemed Garfield's court documents "acceptable in a court of law."
Documented Results
The platform's track record includes widely reported outcomes:
- 7,000 GBP debt recovery for 7.50 GBP in professional fees
- Document generation that passed blind quality assessment
- Automated handling of limitation period checks, claim validity assessment, and pre-action letter generation
If unpaid, the system drafts and submits claim forms, replies to admissions or defenses, handles counterclaims, manages document production, prepares trial bundles, and drafts skeleton arguments.
Judicial Endorsement
Lord Justice Birss, deputy head of civil justice, highlighted Garfield's technology as "absolutely at the core of what we can do for access to justice." This judicial backing matters because it signals that courts may be receptive to AI-assisted filings, reducing the risk that such documents face additional scrutiny.
The Market Gap
Garfield targets a specific niche: high-volume, low-value disputes where traditional representation economics break down. Civil justice statistics show small claims take nearly 50 weeks to reach trial on average. The Ministry of Justice reported over 1.73 million county court claims in 2024.
Most of these claims proceed without professional assistance because the amounts at stake do not justify legal fees. A business owed 5,000 GBP cannot economically spend 1,000+ GBP on recovery. The AI model changes this calculation.
Other UK Entrants
CaseCraft AI has secured 550,000 GBP in funding to address similar market gaps. The platform automates document drafting, deadline tracking, and procedural guidance for small claims.
Both services reflect a broader pattern: AI investment flowing toward underserved legal markets that traditional firms cannot profitably address.
Limitations and Concerns
Several factors constrain AI's access-to-justice impact:
Scope: Garfield currently handles debt recovery under 10,000 GBP and is exploring housing disrepair claims. Complex matters, contested facts, and novel legal questions remain outside the scope.
Digital divide: Users must have digital literacy and internet access. The populations most in need of legal assistance often face technology barriers.
Appeals and complications: When matters become procedurally complex, AI systems may reach their limits. The transition from automated to human assistance is not always smooth.
Accountability: If AI-generated documents cause harm, the liability framework remains unsettled. Indemnification and insurance coverage require ongoing development.
The ABA Perspective
The ABA Task Force Year 2 Report found more than 100 documented AI use cases in legal aid settings. The report characterizes generative AI as "beginning to demonstrate real potential to expand access to legal help."
Key developments include:
- Increased productivity at legal aid organizations
- Direct delivery of understandable legal information to self-represented litigants
- AI-assisted document preparation for pro se filers
What AI Cannot Fix
The access-to-justice gap has structural causes that technology alone cannot address:
- Litigation costs beyond legal fees (court filing fees, time off work, childcare)
- Power imbalances between represented and unrepresented parties
- Complexity of substantive law that resists simplification
- Emotional and psychological barriers to pursuing claims
AI addresses the professional fee component of access-to-justice barriers. Other barriers require policy interventions beyond technology.
Measuring Impact
Meaningful access-to-justice metrics should track:
- Volume of matters handled that would otherwise go unrepresented
- Outcomes compared to traditional representation
- User satisfaction and comprehension
- Error rates and correction processes
- Cost per resolved dispute
Garfield and similar services are generating these data points. The evidence base for AI's access-to-justice impact is growing but remains limited.
The Honest Assessment
AI-powered legal services are not hype when applied to appropriate use cases. A 7.50 GBP letter before action is genuine value for users who previously had no economically viable option.
But AI is not a comprehensive solution to the access-to-justice crisis. It addresses one barrier -- professional cost in repeatable workflows. It does not erase complexity, fear, procedural friction, or the power imbalance between represented and unrepresented parties.
The honest framing is straightforward: AI can widen access at the margins where cost is the primary obstacle and the legal task is structured enough to standardise. That is valuable. It is also limited. Both things can be true at once.
Key Takeaways
- Garfield charges 7.50 GBP for letters before action vs 1,000+ GBP for traditional legal services
- SRA authorization marks the first AI-only law firm in the UK
- Civil justice statistics show 1.73 million county court claims annually, most without professional assistance
- ABA Task Force documents 100+ AI use cases in legal aid settings
- AI addresses cost barriers but cannot solve structural access-to-justice issues like digital divide or substantive complexity