A video can look convincing, sound authentic and still be entirely fabricated.
That problem is becoming increasingly important as generative AI makes it easier to create realistic synthetic media capable of imitating a person’s face, voice or behaviour.
Dubai is now responding to that challenge with a new artificial intelligence model designed to identify signs of video manipulation.
The Dubai Electronic Security Centre has announced SARAAB, an AI-powered system intended to detect deepfake videos by analysing facial characteristics and identifying indicators of hidden or synthetic layers.
The technology is significant not simply because of what it can detect.
It reflects a much broader legal challenge:
As synthetic media becomes harder to distinguish from genuine content, proving whether digital material is authentic may become as important as proving what it appears to show.
Why Deepfakes Are Becoming a Legal Problem
Deepfakes were once largely associated with entertainment or online experimentation.
That has changed.
Synthetic video and audio can now potentially be used for:
- impersonation;
- fraud;
- reputational attacks;
- blackmail;
- identity theft;
- social engineering;
- misinformation; and
- fabrication of supposed evidence.
The legal risk arises because people naturally place significant weight on what they can see and hear.
A realistic video showing a person apparently making a statement or performing an act can therefore cause damage before anyone has determined whether the content is genuine.
Deepfakes and Financial Fraud
One of the clearest risks is fraud.
A convincing synthetic video or voice recording can imitate:
- a chief executive;
- company director;
- bank representative;
- client;
- family member; or
- government official.
The person receiving the communication may then be instructed to:
- transfer money;
- disclose confidential information;
- approve a payment;
- provide authentication credentials; or
- take another action they would ordinarily regard as legitimate.
This changes the nature of traditional impersonation scams.
Previously, suspicious spelling or an unfamiliar email address might provide warning signs.
AI-generated voice and video can add a much stronger appearance of authenticity.
Businesses therefore need to reconsider whether seeing or hearing a known individual is, by itself, sufficient verification for a sensitive transaction.
Deepfakes Can Damage Reputation Almost Instantly
Synthetic media also creates substantial reputational risk.
A fabricated video may appear to show a person:
- making offensive comments;
- engaging in unlawful conduct;
- admitting wrongdoing;
- behaving inappropriately; or
- endorsing something they have never supported.
Once circulated online, such content can spread faster than the affected person can disprove it.
Even where the video is later shown to be false, reputational damage may already have occurred.
This creates potential overlap with areas including defamation, privacy, cybercrime, impersonation and unlawful use of another person’s image or identity.
The fact that the underlying content is artificial does not make the damage artificial.
Can a Deepfake Be Used as Evidence?
Synthetic media also raises important questions for litigation and criminal investigations.
Video and audio evidence can be highly persuasive because it appears to provide a direct record of events.
Deepfake technology complicates that assumption.
Where authenticity is disputed, courts and investigators may increasingly need to consider matters such as:
- the original source of the file;
- metadata;
- chain of custody;
- compression history;
- editing traces;
- device records;
- forensic analysis; and
- corroborating evidence.
A video should not necessarily be accepted as authentic simply because it appears realistic.
Equally, a genuine video should not automatically become unreliable merely because deepfake technology exists.
The issue is verification.
Detection Tools Can Assist, but They Are Not Infallible
SARAAB represents an important technological response to synthetic media.
Its purpose is to identify characteristics suggesting that a video may have been manipulated or artificially generated.
But deepfake detection should not be treated as a simple binary test capable of resolving every dispute.
Synthetic media techniques continue to evolve.
Detection may be affected by:
- video compression;
- resolution;
- editing;
- re-recording;
- social-media processing;
- generation method; and
- the quality of the original material.
That means technical tools should generally form part of a wider evidential assessment rather than operating as the sole basis for a legal conclusion.
The UAE Already Has Laws Relevant to Deepfake Misuse
There may not always be a need for a law carrying the word “deepfake” in its title for synthetic-media conduct to create legal exposure.
Depending on the circumstances, existing UAE laws can already become relevant where manipulated content is used for unlawful purposes.
Potential issues may arise where deepfakes are used to:
- impersonate another person;
- obtain money dishonestly;
- invade privacy;
- threaten or blackmail;
- publish defamatory material;
- misuse personal data;
- access systems or accounts unlawfully; or
- deceive another party.
The applicable legal consequences depend on what the synthetic material is used to achieve.
The technology itself is not necessarily unlawful.
Its use may be.
Identity and Consent Are Becoming More Important
Synthetic media can reproduce a person’s likeness without their participation.
That raises a basic question:
Who controls the digital representation of an individual?
A deepfake may use someone’s:
- face;
- voice;
- name;
- image;
- movements; or
- other identifying characteristics.
Where this is done without consent, particularly for commercial, harmful or deceptive purposes, privacy and personality-right concerns may arise.
Businesses using AI-generated people in advertising should therefore be especially careful where the content resembles or reproduces a real individual.
Employers Should Consider Deepfake Fraud in Their Internal Controls
The development is particularly relevant to corporate compliance.
A company may already require approval from a senior executive before processing a high-value payment.
But what happens if employees receive a realistic video call apparently involving that executive?
Businesses should consider introducing verification procedures that do not depend solely on voice or appearance.
For sensitive instructions, this might include:
- secondary authentication;
- callback procedures;
- predetermined approval channels;
- transaction limits;
- dual authorisation; or
- confirmation through separate secure systems.
The key principle is that visual authenticity should no longer automatically equal identity verification.
Individuals Also Need to Preserve Evidence
Anyone targeted by a suspected deepfake should act quickly.
Useful steps may include preserving:
- the original video or audio file;
- URLs or platform details;
- screenshots;
- timestamps;
- account information;
- messages accompanying the content; and
- evidence showing how and where it was distributed.
Deleting or editing the material may make later forensic analysis more difficult.
Where the content involves fraud, threats, extortion or significant reputational harm, the evidence may become important to any complaint or legal proceedings.
The Burden of Trust Is Changing
For decades, digital media made communication easier.
Generative AI is now forcing people to reconsider how much trust should be placed in digital content.
The old assumption was:
If I can see it, it probably happened.
That assumption is becoming less reliable.
For businesses, courts, regulators and individuals, authenticity increasingly needs to be established rather than assumed.
This is why deepfake-detection technology has significance beyond cybersecurity.
It may become part of:
- fraud prevention;
- evidential verification;
- corporate investigations;
- regulatory enforcement; and
- dispute resolution.
Conclusion
Dubai’s development of SARAAB reflects the growing importance of distinguishing genuine digital evidence from synthetic media.
Deepfakes can create risks extending across fraud, impersonation, privacy, reputation and legal proceedings.
Technology capable of detecting signs of manipulation can assist in addressing those risks, but it does not eliminate the need for human assessment, forensic evidence and proper legal analysis.
As AI-generated media becomes increasingly realistic, the legal question will often move beyond what a video appears to show.
It will also become:
Can we prove that the video itself is real?
Al Kabban & Associates
For businesses seeking guidance, Al Kabban & Associates, with over 30 years of experience in UAE law and recognition by Legal 500, stands ready to help corporations build resilience against legal risks while ensuring compliance with local and international standards. For more information or to schedule a consultation, contact us at +971 4 453 9090 or visit www.alkabban.com. You can also follow us on social media for more updates on everything law related in the UAE: @Alkabban_Law
ALSO READ –
