Ad/Marketing Communication
This legal article/report forms part of my ongoing legal commentary on the use of artificial intelligence within the justice system. It supports my work in teaching, lecturing, and writing about AI and the law and is published to promote my practice. Not legal advice. Not Direct/Public Access. All instructions via clerks at Doughty Street Chambers. This legal article concerns AI Law.
Surridge v The Information Commissioner & The Cabinet Office [2024] UKFTT 597
I’ve been reviewing the decision in Surridge v The Information Commissioner & The Cabinet Office [2024] UKFTT 597 (GRC), available here:
Here are some initial thoughts. While the specific matter concerns “wildfire preparedness,” its broader significance lies in how freedom of information (FOI) requests are handled, particularly involving emerging technologies like artificial intelligence (AI).
Quick Summary of the Facts
In brief, Mr Surridge submitted a Freedom of Information (FOI) request to the Cabinet Office for any documents about “wildfire preparedness for safeguarding people and their homes near woodland.” The Cabinet Office said, “Yes, we have relevant material,” but then argued that finding and extracting it all would exceed the FOI cost limit of £600 (which translates to 24 hours of staff time). The Information Commissioner agreed, and so did the Tribunal in the final decision.
Where AI Steps Into the Picture
Mr Surridge insisted that the Cabinet Office could use AI tools to zip through the records faster, preventing costs from spiralling. The Cabinet Office, however, stated it simply didn’t have the kind of AI software that could do such a heavy lift. Building on earlier cases (like Oakley v IC [2024] UKFTT 315 (GRC)), the Tribunal held that FOI appeals must consider an authority’s actual practices and technology, not hypothetical or desirable ones. So, while AI was central to Mr Surridge’s argument, it didn’t change the ultimate outcome because that technology wasn’t available to the Cabinet Office.
Comment
It’s difficult not to have some sympathy for the appellant’s position. Notably, the FTT declined to summarily dismiss the appeal, recognising the potential utility of AI-generated suggestions:
“[14] The Appellant has provided screen shots of AI suggested search paths that the public authority might consider in interrogating it systems. My reading of those suggestions also provides the Appellant with ways he might refine his information request.
[15] It seems to me that the Tribunal might be assisted by the public authority submissions on its reliance on section 12 FOIA. It would be a matter for the public authority as to whether the AI suggested search pathways should be considered or any submissions on that point.
16. At this stage I do not consider the appeal has no reasonable prospect of success pursuant to Rule 8(3)(c).”
However, there is a balance to be struck between transparency and the practical limits. Even if new tech could, in theory, help locate information more quickly, public authorities aren’t required to implement advanced systems purely to meet FOI requests. The Judge observed:
17. …the Cabinet Office evidence now confirms that these methods are not available to it. The case law set out above essentially says that the Tribunal must assess the applicability of s12 FOIA by looking at the processes actually used and adopted by the public authority.
18. On that basis, and looking at the DN, it is impossible to find an error of law in its conclusion that the Cabinet Office was reasonable to apply s12 FOIA to the request. It was not an error for the Commissioner to accept the account of the Cabinet Office as to how long the searches would take. There is nothing in the suggestion that AI processes could be used (in the light of the evidence that they are not) which could challenge the conclusions reached by the Commissioner.
Looking ahead, as AI becomes more accessible and cost-effective, we could see shifts in how public authorities are expected to handle FOI requests. The Tribunal’s initial cautious openness to AI-generated search suggestions (paragraphs [14]-[16]) indicates potential flexibility in future decisions if public authorities adopt such technologies. This raises intriguing questions about whether public authorities will face increasing expectations to modernise their information retrieval systems, potentially recalibrating the balance between cost, practicality, and the transparency mandated by FOIA.
