UzbekistanEthics Rules for creating, deploying and using AI-based solutions
Order No. 284-mh of the Minister of Digital Technologies of the Republic of Uzbekistan of registered with the Ministry of Justice on under No. 3787 (Ethics Rules for creating, deploying and using solutions based on artificial intelligence)
In force since .
An AI governance rule binding public and private bodies.
- How the hook was established
- express
- What makes it apply
- Place of effect
- Audit expectation
- none
As of .
What it requires
- Do not accept a decision taken by an AI technology or system as the final decision: keep it under human oversight and review and have a person take the final decision.
- Do not rely only on the conclusions of information resources created with AI or of AI-based information systems when taking decisions that bear on a person's rights, freedoms and legitimate interests or decisions in health care.
- Make the algorithms of your AI technology or system transparent and understandable, and prevent possible bias.
- Tell users the rules of use, what the system may and may not be used for, and that AI-based solutions or decisions are not always correct and their results may be used as advice or information only, and write and publish a user guide.
- Give open information on the capabilities, limits and possible risks of your AI technology or system.
These lines are LexLint's reading of the law's text, as of .
What it makes you log
Logging duty
The Ethics Rules set principles and duties for AI participants. Read in full, none of their 19 points names a log, a record, an event-recording capability or an audit trail, and none requires anyone to show or produce on demand something only a record could show.
- Kind
- None
- As of
What this law does
Point 1 of the order approves ethics rules for creating, deploying and using solutions based on artificial intelligence. The Ministry of Justice registered the order on under number 3787. The rules define AI participants as the natural persons and legal persons in Uzbekistan that develop, deploy or use artificial intelligence technologies and systems. Those participants must strictly comply with the rules.
Participants must not accept a decision taken by an artificial intelligence technology or system as the final decision. Developers and deployers must keep their technologies and systems under human oversight and review and make sure that a person takes the final decision. Developers and deployers must make the algorithms of their technologies and systems transparent and understandable. Developers and deployers must prevent possible bias in their technologies and systems.
Developers and deployers must give open information on the capabilities, limits and possible risks of their technologies and systems. Developers and deployers must tell users the rules of use, what the system may and may not be used for, and that AI-based solutions or decisions are not always correct and their results may be used as advice or information only. Developers and deployers must write and publish a user guide for their technologies and systems.
Participants must not rely only on the conclusions of information resources created with artificial intelligence or of AI-based information systems when taking decisions that bear on a person's rights, freedoms and legitimate interests or decisions in health care. The order takes effect three months after its official publication. The national legislation database published the order on . The lex.uz record gives as the date the order took effect.
When LexLint raises it
When your app profile says your app generates content with AI, deploys a chatbot, makes high-risk automated decisions or trains models.