July 21, 2026
As artificial intelligence (“AI”) becomes embedded into everyday business activity, its legal significance will increasingly be felt in disputes, investigations and contested decision-making.
Courts and tribunals are beginning to encounter evidence that has been generated, summarised or influenced by AI systems, and the same underlying issues also affect disclosure, confidentiality, privilege and allegations that reasonable skill and care were not exercised.
Whilst there is currently no AI-specific legislation in Jersey, existing legal principles relating to evidence, disclosure, confidentiality, professional responsibility and procedural fairness will continue to apply. That creates a set of practical risks for organisations, professionals and advisers who use AI without proper safeguards or sufficient understanding of how outputs were produced.
AI-generated documents, summaries or analyses may be challenged on the basis that they do not reflect genuine contemporaneous reasoning or human judgment.
Where AI tools have been used, parties may need to explain:
Questions of provenance are also likely to become more important. If AI has been used to draft correspondence, prepare chronologies or shape witness accounts, parties may need to preserve original source documents and be ready to show that the final material still reflects the witness's own evidence or the party's own position.
Material that appears polished but cannot be traced back to a reliable source may attract particular scrutiny.
The use of AI to process or summarise large volumes of documents also raises questions about disclosure.
In litigation involving large data sets, document review platforms have already begun using AI models to read across collections and identify potentially relevant documents for the review team to assess. That may be both sensible and efficient, but it also means opposing parties may seek information about:
For that reason, organisations should think now about dispute-readiness. Keeping a basic record of what AI tools were used, for what purpose and under whose supervision may become important if the process is later challenged.
AI-generated summaries should not be treated as a substitute for the underlying documents, and key forensic judgments should remain with a responsible human reviewer.
These risks are not limited to the admissibility of evidence.
Professionals across legal, fiduciary, financial and advisory services are also beginning to use AI to accelerate research, drafting, analysis and client communications. Used appropriately, that may improve efficiency. Used uncritically, it may create a route to negligence allegations if the individual or organisation cannot show that reasonable skill, care and judgment were exercised.
The danger is not confined to obvious hallucinations. It also includes:
The risk is not simply that AI may produce incorrect information. AI-generated outputs can appear persuasive and authoritative, even where important issues have been overlooked. Ultimately, the question is likely to remain whether reasonable professional judgment was exercised, rather than whether AI was used.
A further risk arises where confidential or privileged material is disclosed in the course of using AI tools.
This issue may arise whenever staff paste documents, emails, witness accounts or factual scenarios into a system without understanding what happens to that information afterwards. Even where a tool is marketed as secure, organisations should understand whether inputs are retained, reviewed, reused for model improvement or transferred outside the expected environment.
Once sensitive material has been shared inappropriately, the legal and reputational consequences may be difficult to unwind.
Businesses and professional firms should identify where AI is already being used in investigations, document review, drafting, legal analysis or internal summaries, and ensure those uses are capable of later explanation.
Source material should be preserved, use of AI should be documented at a basic level where it may later matter, and staff should be given clear rules on the handling of confidential or privileged information.
In practical terms, organisations should consider:
The safest approach is to treat AI as a support tool only and to maintain human oversight of key evidential, strategic and professional judgments.
As AI becomes increasingly integrated into business operations, organisations should ensure that innovation is supported by appropriate governance, oversight and risk management.
If you are unsure about the legal implications of AI within your organisation, taking advice at an early stage may help avoid unnecessary risk and uncertainty. BCR can provide clear, practical guidance tailored to your circumstances. Get in touch today.
Yes, particularly in matters involving large data sets, but the process must still be defensible and subject to appropriate human oversight.
Yes. Sensitive material may be put at risk if staff use tools without understanding how inputs are stored or reused.
The broader risk is that organisations may struggle to defend the reliability of AI-assisted material or show that appropriate professional judgment was exercised.
For Jersey businesses, the safest approach is to treat AI as a supporting tool, not a decision-maker, and to ensure that key decisions remain capable of being explained and justified by reference to human judgment. Disputes involving AI are likely to increase, and organisations that adopt litigation-ready practices now will be better placed to manage both legal and reputational risk.
As artificial intelligence (“AI”) becomes embedded into everyday business activity, its legal significance will increasingly be felt in disputes, investigations and contested decision-making.
Courts and tribunals are beginning to encounter evidence that has been generated, summarised or influenced by AI systems, and the same underlying issues also affect disclosure, confidentiality, privilege and allegations that reasonable skill and care were not exercised.
Whilst there is currently no AI-specific legislation in Jersey, existing legal principles relating to evidence, disclosure, confidentiality, professional responsibility and procedural fairness will continue to apply. That creates a set of practical risks for organisations, professionals and advisers who use AI without proper safeguards or sufficient understanding of how outputs were produced.
AI-generated documents, summaries or analyses may be challenged on the basis that they do not reflect genuine contemporaneous reasoning or human judgment.
Where AI tools have been used, parties may need to explain:
Questions of provenance are also likely to become more important. If AI has been used to draft correspondence, prepare chronologies or shape witness accounts, parties may need to preserve original source documents and be ready to show that the final material still reflects the witness's own evidence or the party's own position.
Material that appears polished but cannot be traced back to a reliable source may attract particular scrutiny.
The use of AI to process or summarise large volumes of documents also raises questions about disclosure.
In litigation involving large data sets, document review platforms have already begun using AI models to read across collections and identify potentially relevant documents for the review team to assess. That may be both sensible and efficient, but it also means opposing parties may seek information about:
For that reason, organisations should think now about dispute-readiness. Keeping a basic record of what AI tools were used, for what purpose and under whose supervision may become important if the process is later challenged.
AI-generated summaries should not be treated as a substitute for the underlying documents, and key forensic judgments should remain with a responsible human reviewer.
These risks are not limited to the admissibility of evidence.
Professionals across legal, fiduciary, financial and advisory services are also beginning to use AI to accelerate research, drafting, analysis and client communications. Used appropriately, that may improve efficiency. Used uncritically, it may create a route to negligence allegations if the individual or organisation cannot show that reasonable skill, care and judgment were exercised.
The danger is not confined to obvious hallucinations. It also includes:
The risk is not simply that AI may produce incorrect information. AI-generated outputs can appear persuasive and authoritative, even where important issues have been overlooked. Ultimately, the question is likely to remain whether reasonable professional judgment was exercised, rather than whether AI was used.
A further risk arises where confidential or privileged material is disclosed in the course of using AI tools.
This issue may arise whenever staff paste documents, emails, witness accounts or factual scenarios into a system without understanding what happens to that information afterwards. Even where a tool is marketed as secure, organisations should understand whether inputs are retained, reviewed, reused for model improvement or transferred outside the expected environment.
Once sensitive material has been shared inappropriately, the legal and reputational consequences may be difficult to unwind.
Businesses and professional firms should identify where AI is already being used in investigations, document review, drafting, legal analysis or internal summaries, and ensure those uses are capable of later explanation.
Source material should be preserved, use of AI should be documented at a basic level where it may later matter, and staff should be given clear rules on the handling of confidential or privileged information.
In practical terms, organisations should consider:
The safest approach is to treat AI as a support tool only and to maintain human oversight of key evidential, strategic and professional judgments.
As AI becomes increasingly integrated into business operations, organisations should ensure that innovation is supported by appropriate governance, oversight and risk management.
If you are unsure about the legal implications of AI within your organisation, taking advice at an early stage may help avoid unnecessary risk and uncertainty. BCR can provide clear, practical guidance tailored to your circumstances. Get in touch today.
Yes, particularly in matters involving large data sets, but the process must still be defensible and subject to appropriate human oversight.
Yes. Sensitive material may be put at risk if staff use tools without understanding how inputs are stored or reused.
The broader risk is that organisations may struggle to defend the reliability of AI-assisted material or show that appropriate professional judgment was exercised.
For Jersey businesses, the safest approach is to treat AI as a supporting tool, not a decision-maker, and to ensure that key decisions remain capable of being explained and justified by reference to human judgment. Disputes involving AI are likely to increase, and organisations that adopt litigation-ready practices now will be better placed to manage both legal and reputational risk.
As artificial intelligence (“AI”) becomes embedded into everyday business activity, its legal significance will increasingly be felt in disputes, investigations and contested decision-making.
Courts and tribunals are beginning to encounter evidence that has been generated, summarised or influenced by AI systems, and the same underlying issues also affect disclosure, confidentiality, privilege and allegations that reasonable skill and care were not exercised.
Whilst there is currently no AI-specific legislation in Jersey, existing legal principles relating to evidence, disclosure, confidentiality, professional responsibility and procedural fairness will continue to apply. That creates a set of practical risks for organisations, professionals and advisers who use AI without proper safeguards or sufficient understanding of how outputs were produced.
AI-generated documents, summaries or analyses may be challenged on the basis that they do not reflect genuine contemporaneous reasoning or human judgment.
Where AI tools have been used, parties may need to explain:
Questions of provenance are also likely to become more important. If AI has been used to draft correspondence, prepare chronologies or shape witness accounts, parties may need to preserve original source documents and be ready to show that the final material still reflects the witness's own evidence or the party's own position.
Material that appears polished but cannot be traced back to a reliable source may attract particular scrutiny.
The use of AI to process or summarise large volumes of documents also raises questions about disclosure.
In litigation involving large data sets, document review platforms have already begun using AI models to read across collections and identify potentially relevant documents for the review team to assess. That may be both sensible and efficient, but it also means opposing parties may seek information about:
For that reason, organisations should think now about dispute-readiness. Keeping a basic record of what AI tools were used, for what purpose and under whose supervision may become important if the process is later challenged.
AI-generated summaries should not be treated as a substitute for the underlying documents, and key forensic judgments should remain with a responsible human reviewer.
These risks are not limited to the admissibility of evidence.
Professionals across legal, fiduciary, financial and advisory services are also beginning to use AI to accelerate research, drafting, analysis and client communications. Used appropriately, that may improve efficiency. Used uncritically, it may create a route to negligence allegations if the individual or organisation cannot show that reasonable skill, care and judgment were exercised.
The danger is not confined to obvious hallucinations. It also includes:
The risk is not simply that AI may produce incorrect information. AI-generated outputs can appear persuasive and authoritative, even where important issues have been overlooked. Ultimately, the question is likely to remain whether reasonable professional judgment was exercised, rather than whether AI was used.
A further risk arises where confidential or privileged material is disclosed in the course of using AI tools.
This issue may arise whenever staff paste documents, emails, witness accounts or factual scenarios into a system without understanding what happens to that information afterwards. Even where a tool is marketed as secure, organisations should understand whether inputs are retained, reviewed, reused for model improvement or transferred outside the expected environment.
Once sensitive material has been shared inappropriately, the legal and reputational consequences may be difficult to unwind.
Businesses and professional firms should identify where AI is already being used in investigations, document review, drafting, legal analysis or internal summaries, and ensure those uses are capable of later explanation.
Source material should be preserved, use of AI should be documented at a basic level where it may later matter, and staff should be given clear rules on the handling of confidential or privileged information.
In practical terms, organisations should consider:
The safest approach is to treat AI as a support tool only and to maintain human oversight of key evidential, strategic and professional judgments.
As AI becomes increasingly integrated into business operations, organisations should ensure that innovation is supported by appropriate governance, oversight and risk management.
If you are unsure about the legal implications of AI within your organisation, taking advice at an early stage may help avoid unnecessary risk and uncertainty. BCR can provide clear, practical guidance tailored to your circumstances. Get in touch today.
Yes, particularly in matters involving large data sets, but the process must still be defensible and subject to appropriate human oversight.
Yes. Sensitive material may be put at risk if staff use tools without understanding how inputs are stored or reused.
The broader risk is that organisations may struggle to defend the reliability of AI-assisted material or show that appropriate professional judgment was exercised.
For Jersey businesses, the safest approach is to treat AI as a supporting tool, not a decision-maker, and to ensure that key decisions remain capable of being explained and justified by reference to human judgment. Disputes involving AI are likely to increase, and organisations that adopt litigation-ready practices now will be better placed to manage both legal and reputational risk.
As artificial intelligence (“AI”) becomes embedded into everyday business activity, its legal significance will increasingly be felt in disputes, investigations and contested decision-making.
Courts and tribunals are beginning to encounter evidence that has been generated, summarised or influenced by AI systems, and the same underlying issues also affect disclosure, confidentiality, privilege and allegations that reasonable skill and care were not exercised.
Whilst there is currently no AI-specific legislation in Jersey, existing legal principles relating to evidence, disclosure, confidentiality, professional responsibility and procedural fairness will continue to apply. That creates a set of practical risks for organisations, professionals and advisers who use AI without proper safeguards or sufficient understanding of how outputs were produced.
AI-generated documents, summaries or analyses may be challenged on the basis that they do not reflect genuine contemporaneous reasoning or human judgment.
Where AI tools have been used, parties may need to explain:
Questions of provenance are also likely to become more important. If AI has been used to draft correspondence, prepare chronologies or shape witness accounts, parties may need to preserve original source documents and be ready to show that the final material still reflects the witness's own evidence or the party's own position.
Material that appears polished but cannot be traced back to a reliable source may attract particular scrutiny.
The use of AI to process or summarise large volumes of documents also raises questions about disclosure.
In litigation involving large data sets, document review platforms have already begun using AI models to read across collections and identify potentially relevant documents for the review team to assess. That may be both sensible and efficient, but it also means opposing parties may seek information about:
For that reason, organisations should think now about dispute-readiness. Keeping a basic record of what AI tools were used, for what purpose and under whose supervision may become important if the process is later challenged.
AI-generated summaries should not be treated as a substitute for the underlying documents, and key forensic judgments should remain with a responsible human reviewer.
These risks are not limited to the admissibility of evidence.
Professionals across legal, fiduciary, financial and advisory services are also beginning to use AI to accelerate research, drafting, analysis and client communications. Used appropriately, that may improve efficiency. Used uncritically, it may create a route to negligence allegations if the individual or organisation cannot show that reasonable skill, care and judgment were exercised.
The danger is not confined to obvious hallucinations. It also includes:
The risk is not simply that AI may produce incorrect information. AI-generated outputs can appear persuasive and authoritative, even where important issues have been overlooked. Ultimately, the question is likely to remain whether reasonable professional judgment was exercised, rather than whether AI was used.
A further risk arises where confidential or privileged material is disclosed in the course of using AI tools.
This issue may arise whenever staff paste documents, emails, witness accounts or factual scenarios into a system without understanding what happens to that information afterwards. Even where a tool is marketed as secure, organisations should understand whether inputs are retained, reviewed, reused for model improvement or transferred outside the expected environment.
Once sensitive material has been shared inappropriately, the legal and reputational consequences may be difficult to unwind.
Businesses and professional firms should identify where AI is already being used in investigations, document review, drafting, legal analysis or internal summaries, and ensure those uses are capable of later explanation.
Source material should be preserved, use of AI should be documented at a basic level where it may later matter, and staff should be given clear rules on the handling of confidential or privileged information.
In practical terms, organisations should consider:
The safest approach is to treat AI as a support tool only and to maintain human oversight of key evidential, strategic and professional judgments.
As AI becomes increasingly integrated into business operations, organisations should ensure that innovation is supported by appropriate governance, oversight and risk management.
If you are unsure about the legal implications of AI within your organisation, taking advice at an early stage may help avoid unnecessary risk and uncertainty. BCR can provide clear, practical guidance tailored to your circumstances. Get in touch today.
Yes, particularly in matters involving large data sets, but the process must still be defensible and subject to appropriate human oversight.
Yes. Sensitive material may be put at risk if staff use tools without understanding how inputs are stored or reused.
The broader risk is that organisations may struggle to defend the reliability of AI-assisted material or show that appropriate professional judgment was exercised.
For Jersey businesses, the safest approach is to treat AI as a supporting tool, not a decision-maker, and to ensure that key decisions remain capable of being explained and justified by reference to human judgment. Disputes involving AI are likely to increase, and organisations that adopt litigation-ready practices now will be better placed to manage both legal and reputational risk.