AI is flooding Britain’s employment courts with lawsuits

Britain’s employment tribunal system faces an overwhelming rise in claims. A key driver is workers using free generative AI tools like ChatGPT to write their legal filings. This has caused a hundredfold increase in urgent interim relief applications. The backlog of unresolved cases has grown by 55 percent, now at 64,000. Officials paint a stark picture where accessible AI strains the system, delaying justice for genuine claims and raising employer costs.

Key takeaways

• Urgent interim relief applications in UK employment courts have surged a hundredfold.
• The overall backlog of unresolved tribunal cases has jumped 55% to 64,000.
• The crisis is largely driven by workers using free AI (e.g., ChatGPT, Grok) to draft claims, which are often hundreds of pages long and contain fabricated laws.
• Employers face higher costs responding to the flood of claims, whether valid or not.
• Upcoming legislative changes, like the new Employment Rights Act, may worsen the situation by adding new grounds for claims and removing compensation caps.

Unprecedented Surge in Claims and Systemic Backlog

Interim relief applications at British employment courts have risen a hundredfold. Tribunal presidents Barry Clarke and Susan Walker detailed this spike in a memo. It forms part of a wider crisis, with total claims increasing 39 percent in the year to March 2026.

Rising volume threatens the system’s stability. People often use AI such as ChatGPT or Grok to write legal claims at no cost, avoiding lawyer fees. These AI-produced documents frequently span hundreds of pages and cite invented statutes.

Mounting Backlog

The flood of submissions has pushed the unresolved case count up by 55 percent, hitting 64,000. This scenario acts like a shared-resource problem amplified by AI. Workers with real complaints now endure extended waits for resolution. Companies also bear greater expenses when answering claims, even those that are groundless.

The Role of Generative AI in Fueling the Crisis

More workers now turn to AI tools like ChatGPT or Grok to prepare their legal cases without charge. This trend adds major pressure to an already strained system.

Volume and Quality of Filings

Many AI-created legal submissions run to several hundred pages. They commonly include made-up laws and unreasonable requests, posing serious difficulties for the courts.

The memo from tribunal presidents Barry Clarke and Susan Walker quantifies this rise. It shows claims grew 39 percent in the year to March 2026. The pile of unresolved cases increased 55 percent to 64,000. This overload creates a shared-resource dilemma for the entire judicial framework.

Consequences for the Justice System and Stakeholders

People with legitimate employment disputes now confront much longer delays for justice. The stockpile of unfinished tribunal cases rose 55 percent to 64,000 in the year to March 2026.

Businesses feel the impact too. They spend more money responding to the wave of claims, valid or not. Numerous AI-written filings are extraordinarily lengthy, demanding considerable time and resources to counter.

A Grim Picture Painted by Officials

A memo from tribunal presidents Barry Clarke and Susan Walker outlines a system under severe stress. The Economist calls this a shared-resource problem made worse by AI, showing how free legal drafting tools burden the courts collectively. Coming law changes add to this pressure; the new Employment Rights Act creates about 25 new reasons to file a claim and eliminates award limits, potentially increasing case numbers.

Broader Context and Potential Future Developments

AI-assisted legal claims create problems beyond Britain. The United States sees comparable issues, where a federal judge called these lawsuits an existential danger to the court system.

Legislative Changes May Intensify the Flood

In Britain, the forthcoming Employment Rights Act from the Labour government will likely make matters worse. This law introduces approximately 25 new bases for claims and scraps current award ceilings, which may encourage more people to submit cases.

A Contrasting Model from Pakistan

Research from Pakistan points to a possible alternative for AI in courts. Judges there, supplied with AI tools and correct training, managed a larger caseload more efficiently. This shows technology might reduce backlogs when introduced with proper support.