Civil Law And Emotion Aware Contracts And Behaviour Adaptive Legal Obligations In Uae

1. Introduction

The concept of an Emergent Legal Order from Networked Human–Machine Systems concerns the development of legal relationships, obligations, rights, responsibilities and liability through interactions between humans, artificial intelligence (AI), autonomous machines, digital platforms, connected devices and algorithmic networks.

In traditional civil law, legal relationships generally arise through identifiable human or corporate actors. For example, a person enters into a contract, a company performs a service, or an individual causes damage through negligence.

However, networked human–machine systems create more complicated situations in which:

Humans make decisions with assistance from AI systems.

Autonomous machines perform activities without continuous human supervision.

Multiple AI systems exchange information and make interconnected decisions.

Smart contracts automatically execute contractual obligations.

Digital platforms coordinate transactions between numerous participants.

Connected devices generate evidence and influence legally significant decisions.

Responsibility is distributed among developers, manufacturers, operators, owners and users.

In the UAE, these developments must be examined through existing civil-law principles, particularly contractual responsibility, civil liability, causation, fault, compensation, good faith and legal personality.

Central legal proposition: An emergent legal order does not automatically create a new category of legal persons. Rather, it creates new forms of legally significant relationships that must generally be attributed to recognised natural or legal persons under UAE law.

2. UAE legal framework

The UAE does not currently have a single comprehensive civil-law statute establishing an independent legal order for networked human–machine systems. Instead, several laws operate together.

A. Federal Decree-Law No. 25 of 2025 – Civil Transactions Law

In force from 1 June 2026

This is particularly important because the former Federal Law No. 5 of 1985 was replaced by the new Civil Transactions Law.

The new framework governs the fundamental private-law relationships relevant to networked systems:

Legal capacity and the recognition of natural and legal persons.

Formation and performance of contractual obligations.

Civil responsibility for harmful conduct.

Compensation for legally recognised damage.

Interpretation of contractual relationships.

Good faith and the exercise of civil rights.

Ownership, custody and transfer of property.

Allocation of responsibility among participants in transactions.

Application to human–machine networks: Where an AI system concludes a transaction, processes instructions or causes financial damage, the court must identify the legally responsible person or organisation behind the system's design, deployment, operation or use.

The machine's autonomous behaviour does not, by itself, eliminate the responsibility of its human or corporate participants.

B. Federal Decree-Law No. 46 of 2021 – Electronic Transactions and Trust Services

This legislation supports the legal recognition of electronic transactions, electronic signatures, electronic documents and trust services.

Its importance includes:

Recognition of electronic contractual processes.

Authentication of electronic signatures.

Establishing the reliability of electronic records.

Supporting transactions concluded through digital platforms.

Recognition of electronic communications as legally relevant evidence.

A networked machine can therefore participate in the technical execution of a transaction without necessarily becoming the legal contracting party.

C. Federal Decree-Law No. 35 of 2022 – Evidence Law

The Evidence Law is essential when disputes involve machine-generated records.

Electronic evidence includes:

Electronic records.

Electronic instruments.

Electronic signatures and seals.

Emails and electronic correspondence.

Modern means of communication.

Other electronically generated or stored evidence.

The law provides a framework for evaluating the evidential value of such material.

For example, where an autonomous trading system executes an unauthorised transaction, relevant evidence may include system logs, transaction histories, authentication records and communications between the operator and the platform.

D. Federal Decree-Law No. 45 of 2021 – Personal Data Protection Law

Networked human–machine systems frequently process personal information.

This legislation becomes relevant when AI platforms collect, analyse, transfer or use personal data.

Important legal concerns include:

Lawful processing of personal information.

Consent and other applicable legal bases.

Data security.

Rights of data subjects.

Cross-border data transfers.

Accountability of data controllers and processors.

E. Federal Decree-Law No. 34 of 2021 – Combating Rumours and Cybercrimes

This legislation is relevant where networked systems are involved in unlawful digital activity, including unauthorised access, misuse of electronic systems, data interference and other prohibited cyber conduct.

Civil compensation may arise alongside criminal or regulatory consequences where the legal requirements are satisfied.

F. Dubai autonomous vehicle legislation

Dubai's Law No. 9 of 2023 concerning the regulation of autonomous vehicles provides an important sector-specific framework.

It illustrates how the UAE can regulate autonomous technologies through licensing, operational requirements, safety controls and responsibility mechanisms.

It is particularly relevant to disputes involving autonomous vehicles, connected transport infrastructure and human supervision.

3. Meaning of an emergent legal order

An emergent legal order is a system of practical rules, relationships and obligations that develops from the repeated interaction of multiple human and technological participants.

Consider a digital marketplace containing:

A human seller.

A human buyer.

An AI recommendation engine.

An automated payment processor.

A logistics robot.

A smart-contract execution system.

A cloud-service provider.

These participants collectively produce a transaction.

The transaction may be completed without any single human personally controlling every stage.

AI Is Not Taking Over the World It’s Being Integrated by Humans

 

 

bearnetai.com

The resulting legal order has five important characteristics.

3.1 Distributed decision-making

Decisions are produced through interactions among different systems rather than by one identifiable individual.

For example, an AI system may assess a customer's creditworthiness, another system may approve a transaction, and a third system may execute payment.

The legal question is which participant is responsible for the resulting decision and its consequences.

3.2 Continuous contractual relationships

Traditional contracts generally identify parties, obligations and performance requirements.

Networked systems can generate repeated automated transactions through pre-agreed rules.

For example, an industrial AI system may automatically purchase replacement components whenever sensor readings indicate that inventory has fallen below a specified level.

The resulting transactions must still satisfy the legal requirements for valid contractual formation and authority.

3.3 Machine-generated conduct

A machine may initiate an action without receiving a fresh human instruction.

Examples include:

Automated financial transfers.

Autonomous vehicle movements.

Algorithmic purchasing.

Smart-contract execution.

Automated insurance decisions.

Industrial robotic operations.

The relevant civil-law issue is not simply whether the machine acted independently, but whether its conduct can legally be attributed to a person or organisation.

3.4 Network-based dependence

A system may depend on several independent service providers.

For instance, an autonomous delivery operation may rely on:

A vehicle manufacturer.

An AI software developer.

A mapping provider.

A telecommunications company.

A fleet operator.

A maintenance contractor.

A failure in one component may cause damage throughout the network.

3.5 Adaptive and evolving rules

Machine-learning systems can change their outputs through training, feedback and changing environmental information.

This creates a difficult question: who bears responsibility when a system behaves differently from the behaviour anticipated when it was deployed?

The answer may involve inadequate testing, defective design, insufficient monitoring, negligent operation or failure to update the system.

4. Legal personality and the status of machines

One of the most important questions is whether an autonomous machine or AI network can become an independent legal person.

Under the general structure of UAE civil law, legal personality belongs to recognised natural and legal persons. There is no general rule granting every AI system independent legal personality merely because it can make decisions or perform transactions.

This distinction is essential.

ParticipantGeneral legal position
Natural personRecognised legal personality
Registered companyRecognised legal personality
AI softwareNo general independent legal personality
Autonomous robotNo general independent legal personality
Digital platform companyLegal personality through its recognised corporate form
Network of AI agentsNo automatic collective legal personality
Smart contractMechanism for executing agreed obligations, not automatically a legal person

Why machine personality is problematic

Granting independent personality to an AI network would raise several questions:

Who owns its assets?

Who represents it in litigation?

Who pays compensation?

Can it enter enforceable contracts?

Can it hold insurance?

Who bears responsibility for its training?

Can it be declared insolvent?

Who controls its operations after a court order?

Until legislation establishes an appropriate legal status, the safer approach is to allocate legal responsibility to the relevant human and corporate participants.

Important distinction: Technical autonomy is not the same as legal autonomy.

5. Civil liability in networked human–machine systems

Civil liability is the central mechanism through which the UAE legal system can respond to damage caused by interconnected technologies.

The principal question is:

When a network of humans and machines collectively causes damage, who must compensate the injured person?

The answer depends on the nature of the relationship, the applicable statutory provisions, contractual obligations, attribution, causation and proof.

5.1 Liability of AI developers

Developers may face contractual or civil claims where damage results from matters such as:

Defective software architecture.

Inadequate safety testing.

Failure to correct known vulnerabilities.

Misleading representations about system capabilities.

Failure to implement agreed technical specifications.

For example, a developer supplies autonomous warehouse software that repeatedly gives unsafe instructions to robotic equipment. If the defect causes physical damage, the developer's potential responsibility must be examined under the applicable contractual and civil-liability rules.

5.2 Liability of system operators

Operators may be responsible for:

Improper deployment.

Inadequate supervision.

Failure to maintain systems.

Ignoring safety warnings.

Unauthorised modifications.

Failure to respond to known malfunctions.

An operator cannot automatically avoid responsibility by arguing that the machine independently selected the harmful action.

5.3 Liability of manufacturers

Manufacturers may face claims involving defective hardware, inadequate safety mechanisms, manufacturing defects and failures to meet applicable product requirements.

Where software and hardware operate together, technical evidence may be required to distinguish the source of the defect.

5.4 Liability of platform owners

Platform owners may incur contractual or other legally recognised responsibility where their own acts, omissions or contractual undertakings cause damage.

Examples include:

Incorrect automated payment execution.

Failure to protect customer accounts.

Unauthorised changes to platform rules.

Inadequate cybersecurity.

Failure to fulfil agreed service obligations.

5.5 Liability of users

Users may also be responsible when they:

Provide unlawful instructions.

Deliberately manipulate an AI system.

Ignore mandatory safety procedures.

Misuse autonomous equipment.

Introduce unauthorised software modifications.

The allocation of responsibility depends on the actual conduct and the relevant legal relationship.

6. Distributed liability and causation

Networked systems create problems because several participants may contribute to a single harmful event.

Consider the following example.

Example: Autonomous industrial accident

Europa se suma a la carrera de la IA: acciones que podrían despegar en 2026 - Bolsamania.com

 

 

bolsamania.com

Software developer

Supplies defective decision-making software.

Equipment manufacturer

Supplies a robotic machine with inadequate protective mechanisms.

Factory operator

Fails to install a required safety update.

Network provider

Experiences a communication failure affecting emergency signals.

Autonomous machine

Executes an unsafe movement and injures a worker.

Legal question: Which party's conduct caused the injury, and how should responsibility be allocated?

A court may need to examine:

Whether the machine was defective.

Whether the developer breached its contractual or legal obligations.

Whether the operator failed to exercise appropriate care.

Whether the communication failure was foreseeable.

Whether the damage was directly or indirectly caused by each participant.

Whether any intervening event affects legal causation.

Whether the claimant contributed to the damage.

The claimant must establish the relevant elements of the particular cause of action. Multiple participants do not automatically become jointly liable merely because they belong to the same technological network.

6.1 Direct and indirect causation

The former Civil Transactions Law, Federal Law No. 5 of 1985, expressly addressed direct and consequential harm in Articles 282–284.

Those provisions are important to understanding historical UAE case law, but the 1985 statute has been replaced with effect from 1 June 2026. Current disputes must be analysed under Federal Decree-Law No. 25 of 2025, subject to its transitional provisions.

The underlying analytical distinction remains useful:

Direct conduct: The conduct immediately producing the injury.

Indirect conduct: Conduct contributing to the injury through another event or mechanism.

For example, a robotic arm may physically injure a worker, while a defective software update may have contributed to the unsafe movement.

The court must examine the applicable statutory rules and the factual causal chain rather than automatically attributing all responsibility to the final machine that physically caused the injury.

6.2 Concurrent responsibility

Where multiple parties contribute to damage, the court may need to determine:

Each party's legal duty.

The extent of its contribution.

The relevant causal connection.

Any applicable statutory or contractual allocation.

The appropriate measure of compensation.

This is especially important for AI networks in which responsibility is distributed across software, hardware, data and operational infrastructure.

7. Contract law and networked machine transactions

Networked systems increasingly create contractual relationships without requiring a human to personally approve each individual transaction.

7.1 Automated contract formation

Suppose a UAE company uses an AI procurement platform.

The platform automatically:

Identifies a shortage of raw materials.

Searches available suppliers.

Selects a supplier according to predetermined criteria.

Sends a purchase order.

Receives electronic acceptance.

Initiates payment.

The legal questions include:

Did a valid offer exist?

Was there acceptance?

Was the platform authorised to act for the company?

Were the contractual terms sufficiently certain?

Was the transaction within the authority granted?

Can the transaction be attributed to the company?

The AI system's technical participation does not eliminate the need to establish legally effective consent and authority.

7.2 Agency and authority

An AI agent may function as a technical instrument through which a company or individual communicates contractual decisions.

Its authority may be determined by:

Express contractual authorisation.

Delegated operational authority.

The parties' established course of dealing.

Applicable agency principles.

Relevant electronic-transaction legislation.

A company may therefore be bound by a machine-generated transaction where the applicable legal requirements for attribution and authority are satisfied.

However, an unauthorised transaction cannot automatically be treated as binding merely because the platform generated it.

7.3 Smart contracts and automatic execution

Smart Contract

 

 

Smart contracts can automatically execute agreed conditions.

For example:

Payment is released after delivery confirmation.

Digital assets are transferred when a specified condition is satisfied.

Insurance payments are initiated after verification of an insured event.

Supply-chain payments are released when sensor data confirms receipt of goods.

The legal distinction is between the software mechanism and the underlying agreement.

A software program may execute an instruction, but the legal validity of the transaction still depends on the applicable requirements governing consent, capacity, authority, lawful purpose and contractual performance.

7.4 Automated contractual mistakes

Suppose an AI purchasing agent accidentally orders 10,000 units instead of 100.

The legal analysis should consider:

Whether the company authorised the purchase.

Whether the supplier knew or should have known of the mistake.

Whether the contractual terms permit cancellation.

Whether a legally recognised defect of consent exists.

Whether the company or supplier contributed to the error.

Whether the contract was subsequently confirmed or performed.

The mere fact that the mistake originated in an algorithm does not automatically invalidate the contract.

8. Data governance and the human–machine relationship

Networked systems depend on continuous data exchange. Consequently, civil-law disputes can arise from improper data collection, processing, disclosure and use.

The UAE Personal Data Protection Law provides an important framework for these relationships.

Principal areas of legal responsibility

AreaPotential legal issue
Data collectionUnlawful or excessive collection
Automated profilingUnlawful processing or failure to respect applicable rights
Data sharingUnauthorised disclosure
Cloud infrastructureSecurity and processing responsibilities
AI trainingUnlawful use of personal information
Cross-border transfersNon-compliance with transfer requirements
CybersecurityInadequate protection of information

A network may involve a controller, processor, cloud provider and AI developer. Their respective responsibilities must be identified according to their actual roles and applicable legal obligations.

Example

A UAE financial platform uses AI to evaluate customers' financial profiles.

The system mistakenly associates one customer's financial information with another customer and automatically rejects a transaction.

Potential legal issues include:

Accuracy of personal information.

Lawfulness of processing.

Security and access controls.

Contractual obligations owed to the customer.

Causation of financial loss.

Availability of correction or compensation.

The fact that an algorithm produced the incorrect result does not itself establish that every participant in the network is liable.

9. Judicial treatment: at least eight important case laws

There is an important qualification for this topic: UAE courts have not yet established a comprehensive doctrine recognising networked AI systems as independent legal persons or creating a special civil-liability regime for collective machine decision-making.

The following authorities are useful because they address electronic contracting, digital evidence, technological transactions, causation and responsibility.

Some are mainland UAE decisions; others are DIFC authorities. DIFC judgments operate within the DIFC's separate common-law judicial framework and should not be represented as binding precedents for all mainland UAE courts.

Case 1. Dubai Court of Cassation – Civil Appeal No. 468 of 2024

Subject: WhatsApp communications and electronic contract formation

Facts and legal issue:

Two individuals exchanged WhatsApp messages concerning a loan of USD 400,000. They did not execute a conventional signed loan agreement.

The dispute concerned whether their electronic communications could establish a legally enforceable contractual arrangement.

Judicial significance:

The Dubai Court of Cassation recognised the legal significance of the WhatsApp communications and examined their authenticity and the identity of the sender.

The decision demonstrates that electronic communications can establish contractual obligations even without a traditional paper agreement.

Application to networked systems:

This authority is important where AI platforms, automated messaging systems or connected digital interfaces communicate contractual offers and acceptances.

The relevant question is whether the communications can reliably be attributed to the relevant contracting party and satisfy the applicable legal requirements.

Case 2. Dubai Court of Cassation – Real Estate Appeal No. 40 of 2025

Subject: Electronic signatures and digitally executed property transactions

Legal issue:

The dispute concerned a sale agreement executed electronically through the Dubai Land Department's electronic system.

Judicial significance:

The decision addressed the evidentiary effect of electronic signatures and the legal significance of electronically authenticated contractual records under the Evidence Law.

Application to networked systems:

This authority supports the recognition of legally effective digital transactions conducted through integrated technological infrastructure.

For example, a machine-mediated property transaction may involve electronic identity verification, digital signatures, automated documentation and government registration systems.

The decision does not establish that an AI system itself becomes a contracting person.

Case 3. Ras Al Khaimah Court of Cassation – Commercial Appeal No. 46/15 JY (2021)

Subject: Computer-generated business records and commercial evidence

Legal issue:

The case concerned the evidentiary value of computer-generated statements and invoices in a commercial dispute between traders.

Judicial significance:

The decision recognised the relevance of electronically maintained business records in establishing commercial obligations.

It illustrates the importance of modern technology in documenting transactions and proving commercial debts.

Application to networked systems:

AI-based accounting platforms and automated enterprise systems may produce records relevant to establishing:

Payment obligations.

Transaction histories.

Delivery records.

Outstanding debts.

Commercial performance.

The reliability and authenticity of the records remain important.

Case 4. Gate Mena DMCC and Huobi Mena FZE v Tabarak Investment Capital Limited – [2020] DIFC TCD 001

Subject: Cryptocurrency custody, contractual obligations and technological risk

Facts:

The dispute arose from a cryptocurrency transaction involving 300 Bitcoin.

Tabarak Investment Capital participated in arrangements involving a digital wallet and custody of the cryptocurrency during the transaction.

The dispute concerned the parties' contractual responsibilities and the loss of digital assets.

Judicial significance:

The court examined contractual formation, the duties associated with custody, the alleged duty of care and the parties' respective responsibilities.

The judgment is particularly important because it demonstrates that digital assets and technologically mediated transactions remain subject to ordinary legal analysis.

Application to networked systems:

Where a machine-controlled network holds or transfers digital assets, responsibility must be assessed by examining the actual contractual relationship, control arrangements and obligations of the relevant participants.

Case 5. Gate Mena DMCC and Huobi Mena FZE v Tabarak Investment Capital Limited – [2023] DIFC CA 002

Subject: Digital assets, contractual formation and civil responsibility

Facts:

The Court of Appeal considered the dispute arising from the earlier cryptocurrency transaction and examined the contractual and liability issues.

Judicial significance:

The Court of Appeal allowed the appeal in part and ordered a retrial on a particular contractual issue.

It considered whether the parties' conduct and performance could establish a binding contractual arrangement, notwithstanding difficulties concerning the earlier agreement.

Application to networked systems:

The case illustrates that contractual relationships can emerge through conduct and performance rather than only through formal written documents.

This is particularly significant for automated platforms where parties may communicate and perform obligations through multiple connected systems.

Case 6. Gate Mena DMCC and Huobi Mena FZE v Tabarak Investment Capital Limited – [2024] DIFC DEC 002

Judgment delivered: 17 June 2026

Subject: Digital Economy Court, digital-asset custody and contractual risk allocation

Legal issue:

Following the Court of Appeal's decision, the Digital Economy Court considered the retrial concerning the contractual obligations associated with the Bitcoin transaction.

Judicial significance:

The Court examined whether Tabarak had assumed strict responsibility for returning the Bitcoin or whether its obligation was to exercise reasonable care.

It concluded that Tabarak was obliged to exercise reasonable care in maintaining control of the Bitcoin, but did not bear strict responsibility for loss occurring without fault on its part.

Application to networked systems:

This is especially relevant to distributed technological networks.

A company that operates a digital wallet, automated transaction platform or custody system should not be assumed to guarantee every outcome merely because it controls part of the technological process.

The scope of responsibility depends on the actual contractual obligations and applicable law.

Case 7. Larmag Holding B.V. v First Abu Dhabi Bank PJSC – [2019] DIFC CFI 054

Subject: Civil liability, harm and causation under UAE law

Legal issue:

The case considered claims involving alleged harmful conduct and the application of Article 282 of the former UAE Civil Transactions Law.

Judicial significance:

The judgment discussed the essential elements of civil liability, including:

Harmful conduct.

Actual damage.

Causal connection between conduct and damage.

It also referred to the interpretation of the former Article 282.

Application to networked systems:

Where a claimant alleges that an AI system caused financial loss, the claimant must establish the relevant legal elements.

Simply demonstrating that an algorithm produced an incorrect result will not necessarily establish the liability of every developer, operator or platform owner.

The judgment concerns the former law and should be used with the necessary caution when analysing claims under the Civil Transactions Law effective from June 2026.

Case 8. Abu Dhabi Commercial Bank PJSC v Shetty & Others – [2022] EWHC 529 (Comm)

Subject: UAE civil liability and the attribution of harmful conduct

Court: English High Court, Commercial Court.

Legal issue:

The proceedings involved allegations of harmful conduct and claims based on UAE civil-liability principles.

The court considered the former UAE Civil Transactions Law, including Articles 282–287.

Judicial significance:

The case provides a useful judicial discussion of harm, direct and consequential causation, and circumstances in which liability may be excluded because damage resulted from an extraneous cause.

Application to networked systems:

This reasoning is useful where damage results from a combination of autonomous machine behaviour, human instructions, external network failures and third-party interference.

It helps frame the distinction between a participant's own conduct and an external cause.

Jurisdictional qualification: This is an English court judgment applying UAE law, not a UAE court precedent.

Case-law note: The first seven authorities include UAE or DIFC judicial decisions, while the eighth is a foreign judgment applying UAE law. The earlier Gate Mena decisions concern the same underlying dispute and should not be counted as eight unrelated factual precedents.

10. Emergent legal order and collective machine governance

An important development is the emergence of collective decision-making through connected human and machine participants.

For example, a UAE financial institution may operate an interconnected network involving:

AI credit assessment.

Automated fraud detection.

Machine-based identity verification.

Algorithmic risk assessment.

Automated payment execution.

Human compliance supervision.

No single component necessarily controls the entire transaction.

The legal system must therefore identify the relationship between each component and the organisation operating the network.

Four emerging legal models

ModelDescriptionLegal concern
Human-controlled AIHuman makes final decisionsHuman responsibility
Machine-assisted governanceAI recommends decisionsDuty of supervision
Autonomous machine networksMachines coordinate actionsAttribution and causation
Hybrid human-machine governanceHumans and machines jointly influence outcomesDistributed responsibility

10.1 Human oversight

Organisations should maintain meaningful human oversight for decisions involving significant financial, property, safety or personal consequences.

10.2 Algorithmic accountability

Organisations should be able to explain how important automated decisions are generated and identify the persons responsible for the systems.

10.3 Technical transparency

System operators should preserve sufficient information about:

Decision-making processes.

Algorithmic changes.

User instructions.

Access permissions.

System failures.

Transaction execution.

10.4 Responsibility allocation

Contracts between developers, operators, manufacturers and service providers should clearly identify their respective responsibilities.

These are recommended governance mechanisms, not a claim that UAE law already imposes an identical universal AI-governance regime.

11. Remedies available under UAE civil law

Where a networked human–machine system causes legally recognised harm, the injured party may seek appropriate remedies depending on the applicable legal basis.

RemedyPractical application
Monetary compensationFinancial loss caused by wrongful conduct
Contractual damagesFailure to perform contractual obligations
Specific performanceEnforcement of an applicable contractual obligation
Contract terminationSerious contractual non-performance where legally justified
RestitutionRestoration of property or benefits where legally available
Injunctive reliefPrevention of continuing unlawful conduct
Data-related remediesAppropriate relief under applicable data legislation
Corrective measuresRectification of defective records or transactions where legally justified

Example

An autonomous financial network mistakenly transfers AED 2 million to an unauthorised recipient.

The affected institution may need to establish:

How the transaction originated.

Whether the transfer was authorised.

Whether authentication controls were defective.

Whether the operator breached its contractual duties.

Whether an external cyberattack caused the loss.

Which parties legally contributed to the damage.

Depending on the circumstances, recovery, compensation and appropriate interim relief may be available.

12. Challenges for UAE civil law

The emergence of networked human–machine systems creates several important legal challenges.

1. Identification of responsible parties

It may be difficult to determine which participant actually caused the damage.

2. Algorithmic opacity

AI systems may not provide easily understandable explanations for their decisions.

3. Distributed causation

Several independent participants may contribute to a single harmful event.

4. Autonomous contractual execution

Machines may execute transactions beyond the immediate expectations of their owners.

5. Cross-border jurisdiction

The developer, platform operator, cloud provider and injured customer may be located in different countries.

6. Evidence preservation

Digital logs, machine records and transaction histories may be deleted, modified or distributed across multiple servers.

7. Insurance and compensation

Traditional insurance arrangements may not adequately address complex losses caused by interconnected autonomous systems.

8. Legal personality

The question of whether sophisticated AI networks should receive a special legal status remains unresolved under the general UAE civil-law framework.

13. Recommendations for developing UAE legal policy

The UAE could strengthen its approach through the following measures.

1. AI responsibility framework: Establish clearer rules for allocating responsibility among AI developers, manufacturers, operators and users.

2. Network liability standards: Develop guidance for situations involving multiple interconnected technological participants.

3. Mandatory audit mechanisms: Require appropriate technical records for high-risk autonomous systems.

4. Human oversight standards: Establish proportionate supervision requirements for important automated decisions.

5. Digital evidence standards: Improve procedures for authenticating machine-generated records and preserving system logs.

6. AI insurance mechanisms: Encourage insurance arrangements covering specified autonomous-system risks.

7. Contractual transparency: Encourage clear contractual provisions concerning automated authority, system errors and allocation of risk.

8. Cross-border cooperation: Strengthen cooperation concerning digital transactions, cybersecurity and technology-related disputes.

9. Specialised judicial expertise: Develop technical expertise within courts and expert institutions dealing with AI-related disputes.

10. Periodic legislative review: Ensure that civil legislation remains capable of addressing changing forms of machine autonomy.

14. Comparative legal perspective

JurisdictionRelevant developmentSignificance
UAECivil Transactions Law 2025 and electronic-transactions frameworkExisting civil-law principles applied to emerging technologies
European UnionAI Act and product-liability frameworkRisk-based AI regulation
United KingdomCommon-law negligence and product-liability principlesHuman and corporate responsibility
SingaporeElectronic transactions and algorithmic contract disputesAutomated contracting and attribution
United StatesProduct liability and autonomous-system litigationDefective technology and causation

The UAE's approach is particularly interesting because it combines a codified civil-law foundation with specialised digital-economy legislation and sector-specific technological regulation.

15. Conclusion

Civil Law and Emergent Legal Order from Networked Human–Machine Systems in the UAE represents an important development in contemporary private law.

The traditional civil-law model assumes identifiable legal persons, contractual relationships and relatively straightforward causal connections. Networked technological environments challenge these assumptions by distributing decision-making, performance and risk among interconnected humans, companies and autonomous systems.

Nevertheless, technological complexity does not automatically remove existing legal responsibility.

The UAE's Civil Transactions Law, Electronic Transactions and Trust Services Law, Evidence Law, Personal Data Protection Law and cybercrime legislation provide important legal foundations for addressing these disputes.

The most important conclusions are:

AI systems do not automatically acquire independent legal personality.

Automated transactions can create legally enforceable obligations.

Developers, operators, manufacturers and users may incur responsibility according to their respective legal duties.

Distributed technological conduct requires careful examination of causation and attribution.

Digital records and machine-generated evidence are increasingly important in civil litigation.

Contractual arrangements must address the authority and risks associated with automated systems.

Future legal reforms may need to develop more specific rules for interconnected autonomous networks.

Ultimately, the UAE's emerging legal approach should ensure that technological autonomy does not become a mechanism for escaping civil responsibility.

16. Important legislation and authorities for further study

Legal instrument / authoritySubject
Federal Decree-Law No. 25 of 2025Civil Transactions Law
Federal Decree-Law No. 46 of 2021Electronic Transactions and Trust Services
Federal Decree-Law No. 35 of 2022Evidence in Civil and Commercial Transactions
Federal Decree-Law No. 45 of 2021Personal Data Protection
Federal Decree-Law No. 34 of 2021Combating Rumours and Cybercrimes
Dubai Law No. 9 of 2023Autonomous Vehicles
Gate Mena v TabarakDigital assets and contractual responsibility
Larmag Holding v First Abu Dhabi BankCivil liability and causation

Research qualification: The Civil Transactions Law of 2025 became effective on 1 June 2026. The older UAE judgments discussed above were decided under earlier legal frameworks. Their principles must be checked against the new legislation, its transitional provisions and the applicable DIFC or ADGM law before being applied to a current dispute. The UAE is a civil-law jurisdiction, and mainland judgments should not be treated as operating under a universal common-law doctrine of binding precedent.

 

LawGratis provides pro bono legal aid and resources to empower communities. Join us in making justice reachable for everyone! Follow our community on Instagram, YouTube, Facebook, and LinkedIn.

LEAVE A COMMENT