Loading Logo

Law and Disorder: how AI in the legal system may affect reputation

June 2026
 by Adam Ispahani

Law and Disorder: how AI in the legal system may affect reputation

June 2026
 By Adam Ispahani

The advent of generative AI and large language models (LLMs) has upended society at large, significantly altering the established patterns of everyday people while also shifting the balance of power within the business world. While the significance of these developments should not be understated, the speed with which they spread through society should also be considered as a valuable source of information. This rate of change has revealed deficiencies across a range of industries, with many unprepared for this technology’s arrival and now facing a monumental catch-up effort. The impact of AI on multiple facets of the legal system is the latest instance of this dynamic worth examining. The rise of chatbots has created a new class of reputational risks, with multiple legal battles currently being waged to determine what remedies affected parties are owed, while the recent deployment of AI tools within law enforcement agencies has the potential to reshape the field as we know it. Although these various proceedings remain in progress, initial developments can provide valuable intelligence for those seeking to protect their digital reputation, whether it be mitigating particular issues or preventing such problems from occurring in the first place.

Defamation by hallucination

Delivering detailed responses to queries in mere seconds, LLM-powered chatbots represent a significant evolution in the digital search landscape and the latest frontier in online reputation management. Although the outputs from these LLMs regularly present sources from which they claim to have derived an answer, chatbots remain vulnerable to ‘hallucinations’ – generated claims that wrongfully represent fiction as fact. Some instances are relatively harmless, such as misstating a person’s place of birth; however, others can result in far more serious consequences. Wolf River Electric, a solar panel contractor from Minnesota, is currently suing Google, claiming that responses from Gemini falsely suggested the company had faced allegations around its trade practices and had made a settlement with local authorities. Google stated that they sought to rectify this issue as soon as they were alerted; however, Wolf River’s executives have expressed concern over the future of their business following a record number of contract cancellations. Similarly, Canadian musician Ashley MacIsaac alleges that Gemini wrongfully stated that he had been convicted of multiple sensitive crimes. Several upcoming performances were cancelled, with the fallout leaving him fearful for his physical safety.

Even though the merits of these cases are yet to be decided, they lay bare the extent of the reputational risks chatbots can pose. Regarding recourse, Google and other AI players have aimed to settle similar cases wherever possible. However, should future aggrieved parties resist these advances and opt for open court, the outcomes of these trials will be precedent setting, providing a clearer guide for companies and individuals as to what responsibility each side bears in these situations, as well as what is required to prove wrongdoing in this novel digital environment. In the interim, it is critical for businesses and prominent individuals to monitor their LLM profiles, allowing for a quick response to this type of problem. Preventing such a problem outright is a tall order given the ever-changing nature of the relevant algorithms and chatbot model releases; however, ensuring that correct information ranks prominently in traditional searches is an effective course of action. As our research has shown, the majority of information in a chatbot response across the most popular models is in line with the results presented by page one of search engines, which underscores the continued importance of maintaining a strong traditional search profile.

From patrol to parole

AI’s seismic impact can now be felt in the world of criminal justice, as law enforcement agencies look to implement these systems in an effort to improve their efficacy. These various deployments come in different forms, from internal programmes that automate administrative tasks to live facial recognition software that can be active across an entire nation. Many have expressed concern about the potential impact on privacy and liberty that these tools pose to the public, arguing that their widespread use constitutes Orwellian overreach. Yet, recent developments highlight a previously underdiscussed possibility of these AI models being used in an inverse way, as the Metropolitan Police announced that hundreds of its own officers were under investigation following the deployment of one such tool operated by AI software provider Palantir. The alleged offences detected by the surveillance programme range from the mundane, such as abuse of the force’s work-from-home policy, to the overtly concerning, including fraud and corruption, none of which had previously been detected by the Met’s Directorate of Professional Standards. As such, there is now a use case that demonstrates AI’s ability to exert influence beyond civil matters, now stretching into a more serious legal realm.

Although it may seem that the impact of this investigation is largely contained to those in policing, its long-term ramifications are likely to be felt far beyond the sector. The Met plans to make further use of AI tools in its broader efforts, with this pilot acting as a bellwether for the software’s prospects when employed in public settings. As the aforementioned investigations progress, more will be known about how this programme operates and what particular data points are interpreted as risk factors. The most instructive cases will be those for officers who are ultimately acquitted,as these will enable an analysis of what factors cause the AI systems to falsely identify wrongdoing. Understanding these nuances in greater detail will allow for better curation of digital profiles to ensure the risk of being ensnared in this type of difficulty is minimised. While it is reasonable to expect that fine-tuning will improve the accuracy of future technology-aided investigations, the current state of AI development has illuminated how some developers prioritise increasing the volume of functions over incremental but critical accuracy gains, which reaffirms the importance of taking proactive steps in the present to prevent reputational damage.

Arrested development

Generative AI tools and LLMs are undeniably impressive and represent major technological advances that are already impacting global society in a profound way. It is also true that they are imperfect, which is primarily due to inaccuracies that can result in similarly significant consequences for those that are affected. With time, these issues should diminish in number; however, it is not clear when this change will become apparent given the differing development priorities of the companies making these tools. As such, it is crucial for both businesses and individuals to take steps to insulate themselves from potential hazards by monitoring how they are perceived by these systems, as well as how such software is deployed in public settings to ensure that any unnecessary pitfalls can be avoided.

Join our newsletter and get access to all the latest information and news:

Privacy Policy.
Revoke consent.

© Digitalis Media Ltd. Privacy Policy.